Three-dimensional graphics today have become so firmly established in our lives that sometimes we don’t even pay attention to its manifestations.

Looking at a billboard depicting the interior of a room or an advertising video about ice cream, watching the frames of an action-packed film, we have no idea that behind all this lies the painstaking work of a 3D graphics master.

3D graphics is

3D graphics (three-dimensional graphics)- this is a special type computer graphics- a set of methods and tools used to create images of 3D objects (three-dimensional objects).

A 3D image is not difficult to distinguish from a two-dimensional one, since it involves creating a geometric projection of a 3D model of the scene onto a plane using specialized software products. The resulting model can be an object from reality, for example a model of a house, car, comet, or it can be completely abstract. The process of constructing such a three-dimensional model is called and is aimed, first of all, at creating a visual three-dimensional image of the modeled object.

Today, based on 3D graphics, you can create a highly accurate copy of a real object, create something new, and bring the most unrealistic design ideas to life.

3D graphics technologies and 3D printing technologies have penetrated into many areas of human activity and bring enormous profits.

3D images bombard us every day on television, in movies, while working with computers and in 3D games, from billboards, clearly representing the power and achievements of 3D graphics.

The achievements of modern 3D graphics are used in the following industries

1. Cinematography and animation- creation of three-dimensional characters and realistic special effects . Creation computer games - development of 3D characters, virtual reality environments, 3D objects for games.
2. Advertising- the capabilities of 3D graphics allow you to advantageously present a product to the market; using 3D graphics you can create the illusion of a crystal-white shirt or delicious fruit ice cream with chocolate chips, etc. At the same time, in reality, the advertised product may have many shortcomings that are easily hidden behind beautiful and high-quality images.
3. Interior design- design and development of interior design also cannot do without three-dimensional graphics today. 3D technologies make it possible to create realistic 3D models of furniture (sofa, armchair, chair, chest of drawers, etc.), accurately repeating the geometry of the object and creating an imitation of the material. Using 3D graphics, you can create a video showing all the floors of the designed building, which may not even have started construction yet.

Steps to create a 3D image

In order to obtain a 3D image of an object, you must complete the following steps

1. Modeling- construction of a mathematical 3D model of the general scene and its objects.
2. Texturing includes applying textures to created models, adjusting materials and making models look realistic.
3. Lighting settings.
4. (moving objects).
5. Rendering- the process of creating an image of an object using a previously created model.
6. Compositing or compositing- post-processing of the resulting image.

Modeling- creation of virtual space and objects inside it, includes the creation of various geometries, materials, light sources, virtual cameras, additional special effects.

The most common software products for 3D modeling are: Autodesk 3D max, Pixologic Zbrush, Blender.

Texturing is an overlay on the surface of a created three-dimensional model of a raster or vector image that allows you to display the properties and material of an object.

Lighting- creation, direction setting and adjustment of lighting sources in the created scene. Graphic 3D editors, as a rule, use the following types of light sources: spot light (divergent rays), omni light (omnidirectional light), directional light (parallel rays), etc. Some editors make it possible to create a volumetric glow source (Sphere light).

A photorealistic image of a 3D scene is a special image of a scene that takes into account shadows cast by objects, as well as phenomena such as reflection and refraction of light.

The program has three different mechanisms for creating photorealistic images. The first one uses the application POV-Ray , the second – built-in technology NVIDIA OptiX , the third uses Embree - ray tracing kernel developed by Intel.

Selecting and adjusting image quality

It is rarely possible to create a successful photorealistic image on the first try. Typically, you will need to create several photorealistic test images to help you adjust the camera position, brightness, and position of the lights, as well as check that the animation is correct. After this, final visualization is carried out.

But creating a photorealistic image can take varying amounts of time, depending on the complexity of the scene and the parameters that determine image quality. Knowing these parameters, on the one hand, helps to avoid unnecessary time spent on trial imaging, and on the other hand, helps to achieve more High Quality final image.

There are various parameters that allow you to change the quality of the resulting photorealistic image.

Mesh quality . This parameter is set in the document parameters (command ST: Document Parameters), and in addition to photorealism, it also affects the quality of display of objects in the 3D window.

In addition, the image quality setting can be accessed using the panel View .

The higher this parameter, the longer it takes to export the scene to POV format, the more random access memory POV-Ray is used and the longer POV-Ray leads preliminary preparation scenes before rendering (Parsing). In this regard, when carrying out preliminary visualization, it is advisable to reduce the quality of the mesh, perhaps even to a minimum. When performing final rendering, it is better to set the mesh quality to maximum.

1. Photorealistic look

This mechanism for generating photorealistic images is based on NVIDIA technologies OptiX. It is designed to generate high-quality photorealistic images taking into account lighting, as well as material properties such as transparency, refractive index, surface properties, etc.

The mechanism allows you to obtain a photorealistic image directly from the T-FLEX CAD environment, providing user-friendly interface control of scene parameters, image generation quality, as well as the ability to save generation results to a file and print. Using this mechanism, you can obtain a photorealistic image not only from 3D models, but also from imported 3D images.

NVIDEA OptiX technology is used to create photorealistic videos when recording disassembly animations in the “3VX: Disassembly” command

The third engine for generating photorealistic images uses Embree, a ray tracing engine developed by Intel.

Embree uses a central processor for its calculations and is characterized by high performance and image quality.

The interface for working with NVIDIA Optix is ​​identical to the interface for working with Embree, so they will be described together below.

Working with the team

To call the option, use the command:

Pictogram	Ribbon
	Tools → Appearance → Photorealism → Photorealistic Look (NVIDIA GPU)
Keyboard	Text menu
<3RV>	Tools > Photorealistic View (NVIDIA GPU)

Pictogram	Ribbon
	Tools → Appearance → Photorealism → Photorealistic View (CPU)
Keyboard	Text menu
	Tools > Photorealistic View (CPU)

After activating the command, a new window appears in which the image is generated.

The quality of the created image largely depends on the number of iterations. Iteration is the calculation of the color of the image pixels. The number of iterations depends on the image size, mesh density, and number of objects.

The number of iterations is displayed at the bottom of the screen.

Depending on the power of the computer, the complexity of the model and established quality image generation process can take from several minutes to several hours.

The toolbar displays options for working with the command.

Print an image. Allows you to print the resulting image.

Save image. Allows you to export the resulting image to files in raster formats *.bmp, *.jpg, *gif, *tiff, *tif, *.png, *.tga. You can give the file a name and specify where it will be stored.

View Options . Allows you to set image generation parameters. More detailed description options are given below.

Lock view parameters. Allows you to fix the direction of view and the scale of the image. Rotation of the model becomes impossible.

Restart generation. Restarts the generation of a photorealistic image, and the current results are reset.

Pause generation. Allows you to temporarily stop image generation. This frees up computer resources spent on this process, which improves performance.

Selecting the quality of the generated image. In the drop-down list you can select one of four image quality values.

Low and medium quality are used for draft images. When choosing this quality, the system automatically calculates the minimum number of iterations required to obtain images with a certain level of “noise”.

To get the most realistic images, you should select High or Maximum quality. At maximum quality, the number of iterations is unlimited.

Selecting the current active camera. Allows you to select one of the cameras present in the 3D scene. The image will be created according to the position of the selected camera.

In addition to the above options, the “Image quality" It can be changed using the drop-down list in the ST window: Document parameters on the “ tab 3D".

The higher the quality, the higher the mesh density. To obtain the most realistic images, it is recommended to set the quality to at least " Increased."

This parameter is especially important if the model has rounded surfaces.

Visually distinguish between images of different quality.

Very rude	Standard	Very high

The process of creating photorealistic images has high requirements for system characteristics. More detailed information about them can be found on our website or in the chapter “Quick Start”.

Image generation can be stopped at any time. The resulting result can be saved on your computer using the option or immediately sent to print using the option.

Result of the operation:

Photorealistic image

Files with examples of creating photorealistic images are in the library "3D Examples 15\Service Tools\Materials and Photorealism».

For ease of use, you can simultaneously display the photorealistic view window and the model window on the screen. To do this you need to use the command “WO: Open a new document window».

In the dialog box that appears, from the drop-down list you must select “Photorealistic look" Using four drop-down lists, you can configure a convenient arrangement of windows on the screen.

Image Options

Fit into window . The option is active only when the "" When the option is enabled, the image of the specified size is completely displayed on the screen. Fixed image size. When activated, allows you to set the size of the created image. This will enable the toolbar option “Lock view parameters". The image size is specified in pixels. The image of the specified size will be created in its entirety, regardless of whether it fits on the screen or not. To obtain high quality images, it is recommended to set the fixed image size as large as possible. Image quality. This option repeats the list of settings from the main panel. The only difference is the ability to set the number of iterations manually by selecting the image quality "custom" and entering the required number in the field. Number of beam reflections. The parameter is important when generating refractions and reflections.

The background and texture settings completely coincide with the standard 3D view parameters of the same name. You can read more about them in the chapter “Working with the 3D View Window».

Ambient luminance factor. Allows you to adjust the brightness of the scene by adjusting the amount of light falling on objects.

The optimal parameters for creating a photorealistic image are set by default.

Examples of photorealistic images

NVIDIA Optix:

Embree:

2. Realistic image

This mechanism uses POV-Ray technology, a program that uses ray tracing. The conditions for image generation are specified in T-FLEX CAD in text form. The POV-Ray application is included in the package. In addition, the application can be downloaded from the corresponding website.

Image in T-FLEX CAD Photorealistic image (POV-Ray)

The photorealistic image is obtained using ray-tracing. To do this, use the POV-Ray application included in the delivery.

It should be noted that the application POV-Ray requires separate installation. To do this, on the installation CD you need to select the file “ povwin36.exe "from the directory "POV-Ray". Installation of POV-Ray is carried out in English. For users unfamiliar with English language, it is recommended to click all the approving buttons ([ Next ], [Yes] or [I Agree ]) in successively appearing dialog boxes.

To obtain a photorealistic 3D image, the scene is exported to POV format using the settings of the current 3D window. Next, the POV-Ray application is automatically launched to generate the resulting image. Once generation is complete, the resulting image can be viewed in the viewing window and, if desired, saved to a file.

When exporting to POV-Ray, textures are applied to objects in the same way as they are displayed in the T-FLEX CAD 3D window. In addition, together with POV-Ray you can use textures of all formats supported by POV-Ray (gif, tga, iff, ppm, pgm, png, jpeg, tiff, sys).

POV-Ray operates in parallel with other systems, i.e. After launching this application, you can continue working in T-FLEX CAD. However, depending on the complexity of the generated image, POV-Ray may take more resources, and then work in T-FLEX CAD will slow down.

Working with the team

To create a photorealistic image, use the “3VY: Create a realistic image" This command is available when the 3D window is active. Before calling the command, you must set the 3D scene to the desired position, set required material operations, light sources (you can use light sources on the camera). When creating a photorealistic image, it is recommended to use perspective projection.

The command is called in the following way:

Pictogram	Ribbon
	Tools → Appearance → Photorealism → Realistic image (POV-ray)
Keyboard	Text menu
<3VY >	Tools > Realistic image (POV-ray)

T-FLEX CAD stores information about the location of the POV-Ray application and checks its presence each time it is accessed.

In the case when POV-Ray is called for the first time, and also if the system cannot find this application, T-FLEX CAD requests the path to it. In this case, a dialog box appears on the screen, using which you need to set the path to the POV-Ray application. Typically the application is located in the following path: "Program Files\POV-Ray for Windows v3.6\bin" The absence of a corresponding directory indicates that the application is not installed (see paragraph “Basic provisions”).

After calling the command, a dialog box appears on the screen.

Width and Height . Set the width and height of the created photorealistic image in pixels. By default, the size of the current 3D window is set.

Color smoothing. Responsible for smoothing the color of the generated image. The value of this parameter must be greater than 0.

The lower this value, the softer the transition from one color to another will look, but in this case, rendering (i.e., image calculation) will take longer. The value of this parameter can be selected from the list or set independently.

POV-Ray uses a special language to describe a 3D scene. With its help, it is possible to set a large number of different characteristics for the surface of the material, as well as for the interior of the material. Therefore, in T-FLEX CAD, the material has special instructions that determine how the material will look when rendered in POV-Ray (command “3MT:Edit materials", button [ POV material ]). When checking the "Use material substitutions", these instructions will be transmitted to POV-Ray. All materials supplied with the system include specific instructions for POV-Ray. In addition to materials, POV will also export additional instructions for the light source (see "Light Source Options", parameter "POV Instructions").

If the checkbox " Use material substitutions» is disabled, instructions automatically generated by T-FLEX CAD will be sent to POV-Ray, based on material properties such as color and reflectivity.

In the 3D window, one or more light sources are assigned to the camera by default. These light sources are oriented relative to the camera and move with it (see the description "3D View Options"). If the checkbox "Export lights on camera» is turned on, these light sources are transferred to POV-Ray.

Save result to. This displays the path to a temporarily generated output file that POV-Ray will use to save the resulting image to bmp format, and T-FLEX CAD to read it. Therefore, if the T-FLEX CAD application is closed before the result is obtained, the image from this file can be viewed later using any other image viewer.

Everything is temporary generated files During the generation process, images are created in the folder specified in the TEMP system variable. After the image is created, all files except the output one are deleted. The output file itself is stored in this folder until a new photorealistic image is created.

Information for users with experience in POV-Ray

Light source parameters. When creating a photorealistic image using conventional light sources, the shadows of objects are very clear, since the light sources are infinitesimal. In reality, this happens very rarely, so shadows are most often smoothed out. The use of diffused light sources makes shadows smoother and improves the quality and realism of the image. In diffuse light sources, instead of one point light source, several point sources shifted relative to each other are used. The more they are shifted, the less clear the shadow will be. The greater the number of point sources a diffuse source has, the higher the shadow blur and the more time it takes to render.

Normal light Diffused light

A diffuse light source in POV-Ray is a set of point light sources. These light sources are placed in the form of a rectangle, oriented in some way relative to the specified center. The number of light sources along each side of the rectangle may be different. In order for a light source created in T-FLEX CAD to become a diffuse light source in POV-Ray, the following must be written in the properties of the light source in the “POV Instructions” field:

area_light<0.035, 0, 0>, <0, 0.035, 0.035>, 5, 5 adaptive 1 jitter

Here, in triangular brackets, the coordinates of the opposite corners of the rectangle relative to the starting point (the point at which the diffused light source is located) are given. "5, 5" is the number of light sources in each direction. In this case, the total number of point light sources is 5x5=25. “adaptive 1 jitter” - additional parameters, including optimization of shadow calculations.

Antialiasing. During normal visualization, gradation and discontinuity of thin lines may occur at the boundaries of objects. Smoothing through additional calculations can reduce Negative influence of these phenomena.

Border jaggies Anti-aliasing enabled

Anti-aliasing is based on rendering parts of the scene with increased resolution. At the same time, the rendering of the scene slows down. Therefore, you should not enable anti-aliasing during the trial rendering stage. But for final rendering it is advisable to enable anti-aliasing.

Diffuse lighting (Radiosity). Conventional rendering takes into account direct illumination, in which only those areas of objects that are directly illuminated by light from the light source are illuminated. However, in the real world, light does not only come from sources. It is also reflected from objects illuminated by direct light. POV-Ray has the ability to enable a mechanism for calculating diffuse lighting, which in some cases helps improve the realism of the image.

Normal lighting Ambient lighting

Due to big amount additional calculations, the use of the diffuse lighting mechanism can lead to a significant slowdown in rendering. Therefore, the use of ambient lighting in test imaging should only be performed at low resolutions.

To enable the diffused lighting mechanism, go to the " Enable "Window" lines Creating a Photorealistic Image» write down the following:

global_settings(

radiosity ( count 500 minimum_reuse 0.018 brightness 0.8))

The meaning of these instructions, as well as additional information regarding the diffuse lighting mechanism, should be found in the POV-Ray application documentation.

Image Resolution. This parameter significantly affects the time spent on visualization. With unchanged image quality, the rendering speed is directly proportional to the area of ​​the resulting image. When testing visualization, you can limit yourself to small resolutions, for example, 320*240.

Additional INI file: When you start the POV-Ray application, a file is created with ini extension, where exported settings are written. If necessary, you can specify other settings and even override those generated in T-FLEX CAD by specifying them in this file. In this case, the name of this file is indicated in the field of this dialog.

Include rows : In the field of this dialog you can insert strings, which are expressions written in POV format, which will be inserted into the exported file.

Explanation: When you run the command, a POV file is created that has the following structure:

● <генерируемые переменные>

● <включаемые строки>

● <экспортированная 3D сцена>.

Generated Variables

The following variables are included in the exported file:

● fAspectRatio – screen width/height. When overriding the Width and Height settings in an additional INI file, you need to override this variable using<включаемые строки>.

● vSceneMin and vSceneMax – cube vertices that limit the 3D scene in 3D space.

● vSceneCenter – the center of the cube.

● fSceneSize – cube diagonal length.

● vCameraPos – camera position.

● vCamera2Scene – vector from vCameraPos to the center of the cube.

● fCamera2Scene – length of the vCamera2Scene vector.

● cBackColor – background color.

These variables can be overridden or used in<включаемых строках>.

For example:

#declare cBackColor<0.1, 0.1, 0.1>

distance fCamera2Scene / 2

rgb<0, 0, 1>

fog_offset vSceneMin . z

fog_alt (vSceneMax . z - vSceneMin . z) / 4

up<0, 0, 1>

redefines the background color and sets a blue fog, depending on the position and size of the 3D scene.

After setting all the necessary parameters to create a photorealistic image, you need to click on the [ button OK ]. Sometimes, when starting POV-Ray, a dialog box may appear "", to launch the application in this case, just click on the [ button OK].

When creating an animation with photorealism enabled in the " command:Animate the model" it is advisable to wait until the first frame is rendered in POV-Ray to make sure that the window " About POV-Ray(tm) for Windows " does not appear and does not interfere with the creation of animation.

After launching POV-Ray, control is transferred to T-FLEX CAD (i.e. you can continue working with it). At the end of image generation or if it is interrupted, the following message appears on the screen:

If you need to view the resulting image, you need to click on the [ button Yes ]. As a result, a viewing window opens, the image of which can be saved to a file. If viewing and saving the resulting image is not required, then click on the [ No ]. In this case, the result of the photorealistic image will be stored in the system directory for some time (until the next photorealistic image is created) TEMP.

Before the image generation is completed, you can run POV-Ray again (the number of such launches is not limited). Then T-FLEX CAD, performing export to POV, at the end of the process of generating the previous image, will produce new launch POV-Ray applications. Thus, a queue of tasks for generating images is implemented, i.e. a new task is launched after the previous one has completed generation.

Examples of photorealistic images of T-FLEX CAD models

Prototypes for photorealism

In a standard installation, there are prototypes specifically designed to quickly create a photorealistic image. To create documents based on these prototypes, you need to call the command “:Create new document based on prototype file", and on the tab " Photorealism » choose one of two prototypes: « Room" or " Flying around an object».

In each of these prototypes, several light sources, a camera and a coordinate system for linking a 3D fragment are pre-created. The position of these elements can be changed at your discretion by moving the corresponding elements in the drawing window. Also in the 2D window there is a small instruction on how to use the prototype.

Typically, working with these prototypes is carried out as follows: a new document is created based on one of the prototypes. A 3D model (at a suitable scale) is inserted into this document as a 3D fragment or 3D image, a photorealistic image of which must be obtained. Next, several trial renderings are performed to determine the appropriate placement of light sources and camera. At the end, a final visualization is carried out.

The settings that need to be specified for trial and final visualization will be discussed below. But first, it is necessary to say about distinctive features each of the prototypes.

Prototype "Room" » is designed to create a static image. In this prototype, the scene consists of a "room", two lights and a camera. In addition, for convenience, a coordinate system for linking a 3D fragment has been created in advance. By default, the two walls and ceiling of the “room” are not visible, but they can be made visible if you uncheck the “Hide ceiling” checkbox in the 2D window.

Prototype " Flying around an object" is intended for both creating a static image and creating photorealistic animation in which the camera moves around the object. The stage consists of a large circular platform, three light sources and a camera. A coordinate system for linking a 3D fragment has been created in advance in the scene. Additionally, camera position is related to expression and depends on the frame in which the scene is located. In the 2D window, you need to set the duration of the animation (that is, the time during which the camera will fly around the object and return to its original location). The scene must be animated using the “frame” variable, taking into account that the number of frames per second is 25.

Example of using a prototype "Flying around an object" is located in the library "3D Examples 15", in the folder "Service tools\ Photorealistic image\ Flight around an object". By opening the file "Scene based on prototype.grb", you need to select the camera in the 3D window " Camera " Next, you need to use the command “AN: Animate model” and perform animation on the “frame” variable from 0 to 250 with steps of 1.

3D modeling and visualization are necessary when manufacturing products or their packaging, as well as when creating product prototypes and creating 3D animation.

Thus, 3D modeling and visualization services are provided when:

• an assessment of the physical and technical features of the product is necessary even before it is created in its original size, material and configuration;
• it is necessary to create a 3D model of the future interior.

In such cases, you will definitely have to resort to the services of specialists in the field of 3D modeling and visualization.

3D models- an integral component of high-quality presentations and technical documentation, as well as the basis for creating a product prototype. The peculiarity of our company is the ability to carry out a full cycle of work to create a realistic 3D object: from modeling to prototyping. Since all work can be carried out in a complex, this significantly reduces the time and costs of searching for performers and setting new technical specifications.

If we are talking about a product, we will help you release a trial series and set up further production, small-scale or industrial scale.

Definition of the concepts “3D modeling” and “visualization”

3D graphics or 3D modeling- computer graphics, combining the techniques and tools necessary to create three-dimensional objects in technical space.

Techniques should be understood as methods of forming a three-dimensional graphic object - calculating its parameters, drawing a “skeleton” or a three-dimensional non-detailed form; extrusion, extension and cutting of parts, etc.

And under the tools are professional 3D modeling programs. First of all - SolidWork, ProEngineering, 3DMAX, as well as some other programs for volumetric visualization of objects and space.

Volume rendering is the creation of a two-dimensional raster image based on the constructed 3D model. At its core, this is the most realistic image of a three-dimensional graphic object.

Applications of 3D modeling:

• Advertising and Marketing

Three-dimensional graphics are indispensable for the presentation of a future product. In order to start production, you need to draw and then create a 3D model of the object. And, based on the 3D model, using rapid prototyping technologies (3D printing, milling, silicone mold casting, etc.), a realistic prototype (sample) of the future product is created.

After rendering (3D visualization), the resulting image can be used when developing packaging design or when creating outdoor advertising, POS materials and exhibition stand design.

• Urban planning

Using three-dimensional graphics, the most realistic modeling of urban architecture and landscapes is achieved - with minimal costs. Visualization of building architecture and landscape design allows investors and architects to experience the effect of presence in the designed space. This allows you to objectively assess the merits of the project and eliminate the shortcomings.

• Industry

Modern production cannot be imagined without pre-production modeling of products. With the advent of 3D technologies, manufacturers have the opportunity to significantly save materials and reduce financial costs for engineering design. Using 3D modeling, graphic designers create three-dimensional images of parts and objects, which can later be used to create molds and prototypes of the object.

• Computer games

3D technology has been used in the creation of computer games for more than ten years. In professional programs, experienced specialists manually draw three-dimensional landscapes, models of characters, animate created 3D objects and characters, and also create concept art (concept designs).

• Cinema

The entire modern film industry is focused on cinema in 3D format. For such filming, special cameras are used that can shoot in 3D format. In addition, with the help of 3D graphics, individual objects and full-fledged landscapes are created for the film industry.

• Architecture and interior design

The technology of 3D modeling in architecture has long proven itself to be the best. Today, creating a three-dimensional model of a building is an indispensable design attribute. Based on the 3D model, you can create a building prototype. Moreover, both a prototype, repeating only the general outlines of the building, and a detailed prefabricated model of the future structure.+

As for interior design, using 3D modeling technology, the customer can see what his home or office space will look like after renovation.

• Animation

Using 3D graphics, you can create an animated character, “make” him move, and also, by designing complex animation scenes, create a full-fledged animated video.

Stages of developing a 3D model

The development of a 3D model is carried out in several stages:

1. Modeling or creating model geometry

We are talking about creating a three-dimensional geometric model, without taking into account the physical properties of the object. The following techniques are used:

• extrusion;
• modifiers;
• polygonal modeling;
• rotation.

2. Texturing an object

The level of realism of the future model directly depends on the choice of materials when creating textures. Professional programs to work with 3D graphics There are practically no limitations in the possibilities for creating a realistic picture.

3. Setting up light and observation point

One of the most difficult stages when creating a 3D model. After all, the realistic perception of the image directly depends on the choice of light tone, brightness level, sharpness and depth of shadows. In addition, it is necessary to select an observation point for the object. This can be a bird's eye view or scaling the space to achieve the effect of being present in it - by choosing a view of the object from a height of human height.+

4. 3D visualization or rendering

The final stage of 3D modeling. It consists of detailing the display settings of the 3D model. That is, adding graphic special effects such as glare, fog, shine, etc. In the case of video rendering, the exact parameters of 3D animation of characters, details, landscapes, etc. are determined. (time of color changes, glow, etc.).

At the same stage, the visualization settings are detailed: the required number of frames per second and the extension of the final video are selected (for example, DivX, AVI, Cinepak, Indeo, MPEG-1, MPEG-4, MPEG-2, WMV, etc.). If necessary, obtain a two-dimensional raster image, the format and resolution of the image are determined, mainly JPEG, TIFF or RAW.

5. Post-production

Processing captured images and videos using media editors - Adobe Photoshop, Adobe Premier Pro (or Final Cut Pro/ Sony Vegas), GarageBand, Imovie, Adobe After Effects Pro, Adobe Illustrator, Samplitude, SoundForge, Wavelab, etc.

Post-production involves giving media files original visual effects, the purpose of which is to excite the minds of a potential consumer: to impress, arouse interest and be remembered for a long time!

3D modeling in foundry

In foundry production, 3D modeling is gradually becoming an indispensable technological component of the product creation process. If we are talking about casting into metal molds, then 3D models of such molds are created using 3D modeling technologies, as well as 3D prototyping.

But casting in silicone molds is gaining no less popularity today. In this case, 3D modeling and visualization will help you create a prototype of an object, on the basis of which a mold will be made from silicone or other material (wood, polyurethane, aluminum, etc.).

3D visualization methods (rendering)

1. Rasterization.

One of the most simple methods rendering. When using it, additional visual effects (for example, the color and shadow of an object relative to the observation point) are not taken into account.

2. Raycasting.

The 3D model is viewed from a certain, predetermined point - from a human height, a bird's eye view, etc. Rays are sent from the observation point that determine the light and shade of the object when it is viewed in the usual 2D format.

3. Ray tracing.

This rendering method means that when it hits a surface, the ray is divided into three components: reflected, shadow and refracted. This actually forms the color of the pixel. In addition, the realism of the image directly depends on the number of divisions.

4. Path tracing.

One of the most complex 3D visualization methods. When using this 3D rendering method, the propagation of light rays is as close as possible to the physical laws of light propagation. This is what ensures the high realism of the final image. It is worth noting that this method is resource intensive.

Our company will provide you with a full range of services in the field of 3D modeling and visualization. We have all the technical capabilities to create 3D models of varying complexity. We also have extensive experience in 3D visualization and modeling, which you can personally verify by studying our portfolio, or our other works not yet presented on the site (upon request).

Brand agency KOLORO will provide you with services for the production of a trial series of products or its small-scale production. To do this, our specialists will create the most realistic 3D model of the object you need (packaging, logo, character, 3D sample of any product, casting mold, etc.), on the basis of which a prototype of the product will be created. The cost of our work directly depends on the complexity of the 3D modeling object and is discussed individually.

Works made using 3D computer graphics equally attract the attention of both 3D designers and those who have a rather vague idea of ​​how it was done. The most successful 3D works cannot be distinguished from real filming. Such works, as a rule, give rise to heated debates about whether it is a photograph or a three-dimensional fake.
Inspired by the works of famous 3D artists, many take up the study of 3D editors, believing that mastering them is as easy as Photoshop. Meanwhile, programs for creating 3D graphics are quite difficult to master, and learning them takes a lot of time and effort. However, even after studying the tools of a 3D editor, it is not easy for a novice designer to achieve a realistic image. Finding himself in a situation where the scene looks “lifeless,” he cannot always find an explanation for this. What's the matter?
The main problem with creating a photorealistic image is the difficulty of accurately simulating the environment. The picture that is obtained as a result of calculation (visualization) in a three-dimensional editor is the result of mathematical calculations according to a given algorithm. It is difficult for software developers to find an algorithm that would help describe all the physical processes of real life. For this reason, environmental modeling falls on the shoulders of the 3D artist himself.
The hardware capabilities of workstations are increasing every day, which makes it possible to use tools for working with 3D graphics even more effectively. At the same time, the arsenal of 3D graphics editor tools is being improved.
There is a certain set of rules for creating a realistic 3D image. Regardless of which 3D editor you work in and the complexity of the scenes you create, they remain the same. Compliance with these requirements does not guarantee that the resulting image will be similar to the photograph. However, ignoring them will certainly cause failure.
Creating a photorealistic image while working on a 3D project alone is an incredibly difficult task. As a rule, those who devote themselves to 3D graphics and work with it professionally, perform only one of the stages of creating a 3D scene. Some know all the intricacies of modeling, others can masterfully create materials, others “see” the correct lighting of scenes, etc. For this reason, when starting to work with three-dimensional graphics, try to find the area in which you feel most confident and develop your talents.
As you know, the result of working in a 3D editor is a static file or animation. Depending on what your final product will be, approaches to creating a realistic image may vary.

Let's start with composition
The location of objects in a 3D scene is of great importance for the final result. They should be positioned in such a way that the viewer is not at a loss when looking at a part of the object that accidentally appears in the frame, but can recognize all the components of the scene at first glance.
When creating a 3D scene, you need to pay attention to the position of objects relative to the virtual camera. Remember that objects that are closer to the camera lens appear larger in size. For this reason, you need to ensure that objects of the same size are on the same line.
Regardless of the plot of a 3D scene, it must necessarily reflect the consequences of some events that happened in the past.
For example, if someone’s footprints lead to a snow-covered house, then, looking at such a picture, the viewer will conclude that someone has entered the house.
When working on a 3D project, pay attention to the overall mood of the scene. It can be conveyed by a well-chosen element of decoration or a certain range of colors. For example, adding a candle to a scene will emphasize the romance of the setting. If you are modeling cartoon characters, the colors should be bright, but if you are creating a monster, choose dark shades.

Don't forget the details
When working on a 3D project, you should always take into account how visible the object is in the scene, how much it is lit, etc. Depending on this, the object should have a greater or lesser degree of detail. The three-dimensional world is a virtual reality, where everything resembles theatrical scenery. If you won't be able to see the back of the object, don't model it. If you have a bolt with a screwed-on screw, you should not model the thread under the nut; if the facade of the house will be visible in the scene, you do not need to model the interior; if you are creating a night forest scene, you should focus only on those objects that are in the foreground. Trees located in the background will be almost invisible in the rendered image, so it makes no sense to model them down to the leaf.
Often, when creating three-dimensional models, small details play almost the main role, making the object more realistic.
If you can't achieve realism in a scene, try increasing the level of detail in the objects. The more fine details the scene contains, the more believable the final image will look. The option of increasing the detail of the scene is almost a win-win, but has one drawback - a large number of polygons, which leads to an increase in rendering time.
You can make sure that the realism of the coupling directly depends on the level of detail at simple example. If you create three models of blades of grass in a scene and visualize them, the image will not make any impression on the viewer. However, if this group of objects is cloned many times, the image will look more impressive.
You can control detail in two ways: as described above (increasing the number of polygons in the scene), or increasing the texture resolution.
In many cases, it makes sense to pay more attention to creating the texture rather than the object model itself. At the same time, you will save system resources required to render complex models, thereby reducing rendering time. It's better to make a better texture than to increase the number of polygons. A great example of clever use of texture is the wall of a house. You can model each brick individually, which will take both time and resources. It's much easier to use a photo of a brick wall.

If you need to create a landscape
One of the most difficult tasks that 3D graphics designers often have to deal with is modeling nature. What is the problem of creating the natural environment around us? The whole point is that any organic object, be it an animal, a plant, etc., is heterogeneous. Despite the apparent symmetrical structure, the shape of such objects does not lend itself to any mathematical description that 3D editors deal with. Even those objects that, at first glance, have a symmetrical appearance, upon closer examination turn out to be asymmetrical. For example, the hair on a person’s head is located differently on the right and left sides; most often it is combed to the right, and a leaf on a tree branch may be damaged by a caterpillar in some place, etc.
The most the best solution To simulate organic matter in three-dimensional graphics, one can consider a fractal algorithm, which is often used in the settings of materials and various three-dimensional modeling tools. This algorithm is better than other mathematical expressions in simulating organic matter. Therefore, when creating organic objects, be sure to use the capabilities of the fractal algorithm to describe their properties.

Subtleties of creating material
The materials that are simulated in 3D graphics can be very diverse - from metal, wood and plastic to glass and stone. Moreover, each material is determined by a large number of properties, including surface topography, specularity, pattern, size and brightness of glare, etc.
When visualizing any texture, you need to remember that the quality of the material in the resulting image highly depends on many factors, including: lighting parameters (brightness, angle of incidence of light, color of the light source, etc.), visualization algorithm (type of visualizer used and its settings ), raster texture resolution. The method of projecting texture onto an object is also of great importance. An unsuccessfully applied texture can “give away” a three-dimensional object as a seam or a suspiciously repeating pattern. In addition, usually real objects are not perfectly clean, that is, there are always traces of dirt on them. If you are modeling a kitchen table, then despite the fact that the pattern on the kitchen oilcloth is repeated, its surface should not be the same everywhere - the oilcloth may be worn at the corners of the table, have cuts from a knife, etc.
To your three dimensional objects without looking unnaturally clean, you can use hand-made (for example, Adobe Photoshop) dirt maps and mix them with the original textures to create a realistic, worn-out material.

Ways to achieve realism in 3D graphics

Works made using three-dimensional computer graphics equally attract the attention of both 3D designers and those who have a rather vague idea of ​​how it was all done. The most successful 3D works cannot be distinguished from real filming. Such works, as a rule, give rise to heated debates about what it is: a photograph or a three-dimensional fake. Inspired by the work of famous 3D artists, many take up learning 3D editors, believing that they are as easy to master as Photoshop. Meanwhile, programs for creating 3D graphics are quite difficult to master, and learning them takes a lot of time and effort. But even after studying the tools of a 3D editor, it is not easy for a beginning 3D designer to achieve a realistic image. Finding himself in a situation where the scene looks “dead,” he cannot always find an explanation for this. What's the matter?

The main problem with creating a photorealistic image is the difficulty of accurately simulating the environment. The picture that is obtained as a result of calculation (visualization) in a three-dimensional editor is the result of mathematical calculations according to a given algorithm. It is difficult for software developers to find an algorithm that would help describe all the physical processes that take place in real life. Therefore, modeling the environment rests on the shoulders of the 3D artist himself. There is a certain set of rules for creating a realistic 3D image. Regardless of which 3D editor you work in and the complexity of the scenes you create, they remain the same. The result of working in a 3D editor is a static file or animation. Depending on what the final product will be in your case, approaches to creating a realistic image may differ.

Let's start with composition

The location of objects in a 3D scene is of great importance for the final result. They should be positioned in such a way that the viewer is not at a loss when looking at a part of the object that accidentally appears in the frame, but can recognize all the components of the scene at first glance. When creating a 3D scene, you need to pay attention to the position of objects relative to the virtual camera. Remember that objects located closer to the camera lens appear larger in size. Therefore, you need to ensure that objects of the same size are on the same line. Regardless of the plot of a 3D scene, it must necessarily reflect the consequences of some events that happened in the past. So, for example, if someone’s footprints lead to a snow-covered house, then, looking at such a picture, the viewer will conclude that someone has entered the house. When working on a 3D project, pay attention to the overall mood of the scene. It can be conveyed by a well-chosen element of decoration or a certain range of colors. For example, adding a candle to a scene will emphasize the romance of the setting. If you are modeling cartoon characters, the colors should be bright, but if you are modeling a disgusting monster, choose dark shades.

Don't forget the details

When working on a 3D project, you should always take into account how visible the object is in the scene, how lit it is, etc. Depending on this, the object should have a greater or lesser degree of detail. The three-dimensional world is a virtual reality, where everything resembles theatrical scenery. If you won't be able to see the back of the object, don't model it. If you have a bolt with a nut screwed on, do not model the threads under the nut; if the facade of the house is visible in the scene, there is no need to model the interior; If you are modeling a night forest scene, you should focus only on those objects that are in the foreground. Trees located in the background will be almost invisible in the rendered image, so it makes no sense to model them down to the leaf.

Often, when creating three-dimensional models, small details play almost the main role, making the object more realistic. If you can't achieve realism in a scene, try increasing the level of detail in the objects. The more fine details the scene contains, the more believable the final image will look. The option of increasing the detail of the scene is almost a win-win, but has one drawback - a large number of polygons, which leads to an increase in rendering time. You can see that the realism of a scene directly depends on the level of detail using this simple example. If you create three models of blades of grass in a scene and visualize them, the image will not make any impression on the viewer. However, if this group of objects is cloned many times, the image will look more impressive. You can control detail in two ways: as described above (increasing the number of polygons in the scene), or increasing the texture resolution. In many cases, it makes sense to pay more attention to creating the texture rather than the object model itself. At the same time, you will save system resources required to render complex models, thereby reducing rendering time. It's better to make a better texture than to increase the number of polygons. A great example of clever use of texture is the wall of a house. You can model each brick individually, which will take both time and resources. It's much easier to use a photo of a brick wall.

If you need to create a landscape

One of the most difficult tasks that 3D designers often have to deal with is modeling nature. What is the problem of creating the natural environment around us? The whole point is that any organic object, be it an animal, a plant, etc., is heterogeneous. Despite the apparent symmetrical structure, the shape of such objects does not lend itself to any mathematical description that 3D editors deal with. Even those objects that, at first glance, have a symmetrical appearance, upon closer examination turn out to be asymmetrical. So, for example, the hair on a person’s head is located differently on the right and left sides; most often he combs it to the right, and a leaf on a tree branch may be damaged by a caterpillar in some place, etc. The best solution for simulating organic matter in 3D can be considered a fractal algorithm, which is often used in the settings of materials and various 3D modeling tools. This algorithm is better than other mathematical expressions in simulating organic matter. Therefore, when creating organic objects, be sure to use the capabilities of the fractal algorithm to describe their properties.

Subtleties of creating material

The materials that are simulated in 3D graphics can be very diverse - from metal, wood and plastic to glass and stone. Moreover, each material is determined by a large number of properties, including surface relief, specularity, pattern, size and brightness of glare, etc. When visualizing any texture, you need to remember that the quality of the material in the resulting image very much depends on many factors, including lighting parameters (brightness, angle of incidence of light, color of the light source, etc.), visualization algorithm (type of renderer used and its settings), raster texture resolution. Also great importance has a method for projecting texture onto an object. An unsuccessfully applied texture can “give away” a three-dimensional object as a seam formed or a suspiciously repeating pattern. In addition, in reality, objects are usually not perfectly clean, that is, there are always traces of dirt on them. If you are modeling a kitchen table, then, despite the fact that the pattern on the kitchen oilcloth is repeated, its surface should not be the same everywhere - the oilcloth may be worn at the corners of the table, have cuts from a knife, etc. To prevent your 3D objects from looking unnaturally clean, you can use hand-made (for example, Adobe Photoshop) dirt maps and mix them with the original textures to create a realistic "worn" material.

Adding motion

When creating animation, the geometry of objects plays a more important role than in the case of a static image. During movement, the viewer can see objects from different angles, so it is important that the model looks realistic from all sides. For example, when modeling trees in a static scene, you can use a trick and simplify your task: instead of creating a “real” tree, you can make two intersecting perpendicular planes and apply a texture to them using a transparency mask. When creating an animated scene, this method is not suitable, since such a tree will look realistic only from one point, and any camera rotation will “give away” the fake. In most cases, once 3D objects disappear from the virtual camera lens, it is best to remove them from the scene. Otherwise, the computer will perform a useless task, calculating invisible geometry.

The second thing that needs to be taken into account when creating animated scenes is the movement in which most objects are in reality. For example, the curtains in the room are swaying in the wind, the clock hands are moving, etc. Therefore, when creating animation, it is necessary to analyze the scene and identify those objects for which it is necessary to set movement. By the way, movement adds realism to static scenes. However, unlike animated ones, in them the movement should be guessed in frozen little things - in a shirt slipping off the back of a chair, a crawling caterpillar on a trunk, a tree bent by the wind. If for more simple objects scenes, creating realistic animation is relatively easy, but simulating a character’s movement without auxiliary tools is almost impossible. In everyday life, our movements are so natural and habitual that we do not think, for example, whether to throw back our heads when laughing or bend down when passing under a low canopy. Modeling such behavior in the world of three-dimensional graphics is associated with many pitfalls, and recreating the movements, and especially facial expressions, of a person is not so easy. That is why, to simplify the task, we use next way: a large number of sensors are attached to the human body, which record the movement of any part of it in space and send a corresponding signal to the computer. He, in turn, processes the information received and uses it in relation to some skeletal model of the character. This technology is called motion capture. When moving the shell, which is placed on the skeletal base, it is also necessary to take into account muscular deformation. For 3D animators involved in character animation, it will be useful to study anatomy in order to better understand the bone and muscle systems.

Lighting is not only light, but also shadows

Creating a scene with realistic lighting is another challenge that must be overcome in order to give the final image greater realism. In the real world, light rays are repeatedly reflected and refracted by objects, resulting in the shadows cast by objects generally having fuzzy, blurred boundaries. The rendering apparatus is mainly responsible for the quality of shadow display. There are separate requirements for shadows cast in a scene. The shadow cast from an object can say a lot - how high it is above the ground, what is the structure of the surface on which the shadow falls, what source illuminated the object, etc. If you forget about shadows in a scene, such a scene will never look realistic, since in reality every object has its own shadow. In addition, a shadow can emphasize the contrast between the foreground and background, as well as “give out” an object that is not in the field of view of the virtual camera lens. In this case, the viewer is given the opportunity to imagine the surrounding environment of the scene. For example, on the shirt of a three-dimensional character, he can see a falling shadow from branches and leaves and guess that there is a tree growing on the opposite side of the shooting point. On the other hand, too many shadows will not make the image more realistic. Make sure that the subject does not cast shadows from auxiliary light sources. If there are several objects in the scene that emit light, for example, lanterns, then all elements of the scene should cast shadows from each of the light sources. However, if in such a scene you will use auxiliary light sources (for example, to highlight dark areas of the scene), there is no need to create shadows from these sources. The auxiliary source should be invisible to the viewer, and the shadows will reveal its presence.

When creating a scene, it is important not to overdo it with the number of light sources. It is better to spend a little time choosing the best position for it than to use several light sources where you can get by with just one. In the case where the use of several sources is necessary, make sure that each of them casts shadows. If you cannot see the shadows of a light source, then perhaps another, stronger source is overexposing them. When arranging light sources in a scene, be sure to pay attention to their color. Daylight sources have a blue tint, but to create an artificial light source you need to give it a yellowish color. It should also be taken into account that the color of the source simulating daylight, also depends on the time of day. Therefore, if the plot of the scene involves evening time, the lighting may be, for example, in the reddish shades of sunset.

The most important thing is miscalculation

Visualization is the final and, of course, the most important stage in creating a three-dimensional scene. The 3D graphics editor calculates the image, taking into account the geometry of objects, the properties of the materials from which they are made, the location and parameters of light sources, etc. If we compare working in 3ds max with video shooting, then the value of the rendering engine can be compared with the film on which the material is shot. Just as two films from different companies can produce bright and faded photographs, the result of your work can be realistic or only satisfactory depending on which image rendering algorithm you choose. The existence of a large number of visualization algorithms has caused an increase in the number of external connected renderers. Often the same renderer can be integrated with different 3D graphics packages. In terms of speed and quality of the rendered image, external visualizers, as a rule, are superior to the standard rendering apparatus of 3D editors. However, it is impossible to give a clear answer to the question which of them gives the best result. The concept of “realism” in this case is subjective, because there are no objective criteria by which one could evaluate the degree of realism of the visualizer.

However, we can say for sure that in order for the final image to be more realistic, the visualization algorithm must take into account all the features of light wave propagation. As we said above, when a ray of light hits objects, it is reflected and refracted many times. It is impossible to calculate the illumination at each point in space, taking into account an infinite number of reflections, so two simplified models are used to determine the light intensity: Raytracing and the Global Illumination method. Until recently, the most popular rendering algorithm was light ray tracing. This method consisted in the fact that a three-dimensional editor tracked the course of a beam emitted by a light source with a given number of refractions and reflections. Tracing cannot provide a photorealistic image because the algorithm does not provide the effects of reflective and refractive caustics (flares resulting from the reflection and refraction of light), as well as light scattering properties. Today, the use of the global illumination method is a prerequisite for obtaining a realistic image. If the tracing calculates only those areas of the scene that receive rays of light, the global illumination method calculates the scattering of light in unlit or shadowed areas of the scene based on an analysis of each pixel in the image. This takes into account all reflections of light rays in the scene.

One of the most common methods for calculating global illumination is Photon Mapping. This method involves calculating global illumination based on the creation of a so-called photon map - information about the illumination of the scene collected using tracing. The advantage of Photon Mapping is that once saved as a photon map, photon tracing results can later be used to create global illumination effects in 3D animation scenes. The quality of Global Illumination calculated using photon tracing depends on the number of photons, as well as the tracing depth. Using Photon Mapping you can also calculate caustics. In addition to calculating global illumination, external renderers allow you to visualize materials taking into account the effect of subsurface scattering (Sub-Surface Scattering). This effect is a necessary condition for achieving realism in materials such as leather, wax, thin fabric, etc. Rays of light falling on such a material, in addition to refraction and reflection, are scattered in the material itself, thereby causing a slight glow from the inside.

Another reason why images rendered using plug-in renderers are more realistic than pictures rendered using standard rendering algorithms is the ability to use camera effects. These include, first of all, depth of field (Depth of Field), blurring of moving objects (motion blur). The depth of field effect can be used when you want to draw the viewer's attention to some detail in the scene. If an image contains a depth of field effect, the viewer will first notice the elements in the scene that are being focused. The depth of field effect can be helpful when you need to visualize what a character is seeing. Using the depth of field effect, you can focus the character's gaze on one or another object. The effect of depth of field is an essential component of a realistic image even when attention in the scene is drawn to a small object - for example, a caterpillar on a trunk. If all the objects that come into focus are equally clearly drawn in the picture, including branches, leaves, trunk and caterpillar, then such an image will not look realistic. If such a scene existed in reality, and the filming was carried out not with a virtual camera, but with a real camera, only the main object would be in focus - the caterpillar. Anything at a distance from it would appear blurry. Therefore, a three-dimensional image must have a depth of field effect.

Conclusion

The hardware capabilities of workstations are increasing every day, which makes it possible to use tools for working with 3D graphics even more effectively. At the same time, the arsenal of 3D graphics editor tools is being improved. At the same time, the basic approaches to creating photorealistic images remain unchanged. Compliance with these requirements does not guarantee that the resulting image will be similar to the photograph. However, ignoring them will certainly cause failure. Creating a photorealistic image while working on a 3D project alone is an incredibly difficult task. As a rule, those who devote themselves to 3D graphics and work with it professionally, show themselves only at one of the stages of creating a 3D scene. Some know all the intricacies of modeling, others know how to skillfully create materials, others “see” the correct lighting of scenes, etc. Therefore, when starting to work with 3D, try to find the area in which you feel most confident and develop your talents.

Sergey and Marina Bondarenko, http://www.3domen.com