Novel View Synthesis of Humans using Differentiable Rendering
This addresses the problem of generating realistic human images from new viewpoints for computer vision and graphics applications, representing an incremental improvement with a novel rendering method.
The paper tackles novel view synthesis of humans by introducing a differentiable renderer using diffuse Gaussian primitives, achieving highly realistic image synthesis from any viewpoint and demonstrating applications like motion transfer and pose re-rendering on Human3.6M and Panoptic Studio datasets.
We present a new approach for synthesizing novel views of people in new poses. Our novel differentiable renderer enables the synthesis of highly realistic images from any viewpoint. Rather than operating over mesh-based structures, our renderer makes use of diffuse Gaussian primitives that directly represent the underlying skeletal structure of a human. Rendering these primitives gives results in a high-dimensional latent image, which is then transformed into an RGB image by a decoder network. The formulation gives rise to a fully differentiable framework that can be trained end-to-end. We demonstrate the effectiveness of our approach to image reconstruction on both the Human3.6M and Panoptic Studio datasets. We show how our approach can be used for motion transfer between individuals; novel view synthesis of individuals captured from just a single camera; to synthesize individuals from any virtual viewpoint; and to re-render people in novel poses. Code and video results are available at https://github.com/GuillaumeRochette/HumanViewSynthesis.