Custom 3D Services

How do we create a 3d video

• Modeling the assets: All the assets you see in the video must be modelled. A simple asset like a table might take 2 hours to do. While a complicated asset like a soldier with clothes and equipment can take a whole week to do.
• Once those assets are modelled. They must be put in the 3D environment inside Unreal engine or any other 3d engine. The assets that move like a horse or a human must be rigged. Rigging is a process where we add virtual bones to a model. Those bones indicate to the 3D engine how each part of the 3d models will move relative to each other. To give an example, an upper arm (radius) will be rigged with a bones that can only do a 180 degree rotation relative to it’s lower part (humerus). This step of the rigging process is called weight painting. We indicate to the 3d engine, what this bone should and shouldn’t do.
• Once the objects are created and rigged, we must then apply materials and textures. Every part of the object must be applied with a material and texture. A combination of different materials and textures can give the impression that this part is metallic, opaque, transparent or any kind of effect we want. We do not create the materials and the textures you see in our videos. We take them from many sources on internet and we edit it with substance painter and substance designer. Once we have a material that satisfy us, we import it in Blender and we apply it to the model.
• Once the models are rigged and have their materials, we must animate them. This is the hardest part to do. To animate them, animators have many ways they can go about. Using the bones they created in the rigging process, they can move them, frame by frame, to give the impression that the model is running talking or jumping. This method of doing thing is complicated because each frame must be done correctly, otherwise it will not look natural. The other way is to use motion capture (motion capture). This method requires to use a living human model and a camera. In the past, it was also required to have a green screen but with a.i, it is no longer necessary. Depending on the software you use, we will assign every bone of the virtual character to a part of the human model. With the help of a.i, the software will recognize the corresponding bone of the human model. While the human model will be moving, the software will analyse his movement and reproduce it with the virtual model. At Netfrater, we use both models, depending of the complexity of the animation and many other factors.
• Once all the animations are finished, we need to put a camera inside the 3d scene or level. Inside a timeline, we will make a camera move and rotate while the characters are doing their animations. If there is 20 scenes in the video, we will have to do this process 20 times. If the scenes happen in different environment, we must load those new environments and redo the process.

• Number of assets (models, objects, characters)
• Number and variety of environment
• Number of cuts and scenes
• Number and complexity of animations

• Dialogues where the mouth and tongues must move according to a script
• Facial animations