With the use of its proprietary AI model, OPPO has produced dynamic, advanced avatars for the first time, providing users with a digital life that is rife with opportunities. OPPO has introduced its Structured Local Radiance Fields for Human Avatar Modeling, which is effective in both learning how people look with loose clothing and modelling skeleton motion.
In practice, the AI model analyses RGB data to create 3D, lifelike, and animated digital avatars with graceful dancing moves using just a few smartphones. People can now access more opportunities in the digital world than ever before.
How does OPPO achieve its human avatar modelling?
At the OPPO Digital Avatar Lab, OPPO has used a few Find X5 Pro smartphones that are used to gather data about the human body. After the recording is done, AI uses RGB videos to analyze and create a 3D model of the person. This gets rid of the problems with reconstruction that older methods had.
AI automatically creates the 3D human model by examining RGB videos when the recording is complete. Anchored as pre-defined nodes, the movements of a dancer’s skeleton are used to build a set of organized local radiance fields. This gives the image a more realistic look and makes it easier to animate. This approach was shown at the Computer Vision and Pattern Recognition Conference (CVPR).
The dancer’s skeleton creates structured local brightness fields. Each local radiance field shows the shape and appearance of the area around its node. This is how fabric deformations are broken down into movements of the skeleton. Through residual movements, this method improves complex clothing deformations.
This lets the avatar dance based on bodily positions. The model can handle more possibilities for future interaction sessions. Neural scene rendering encodes and learns dancing routines using neural networks, generating a dynamic digital character in hours. The girl can eventually dance with her floral dress swaying.
3D, realistic, animated, a breakthrough of OPPO AI technology
OPPO’s patented technology increases neural radiance fields and dynamic character models. The programme uses structured local radiance fields associated with preset places on the human body. This technique trains models to understand brands, hemlines, wrinkles, and fabric texture, even for new positions.
The new Human Avatar Modeling generates 3D, 1:1 digital avatars with better realism and convenience. The method removes restrictions and lets people express themselves online. OPPO makes real improvements and applies the method to various elements of people’s lives, such as virtual apparel fitting, AI fitness training, and lifelike avatars in VR/AR worlds. With one body design, game makers may create more realistic characters and animations.