Convolutional Generation of Textured 3D Meshes
NIPS 2020, 2020.
Our approach can be enriched by employing different forms of supervision as well as incorporating more conditional information that would allow the model to disentangle further aspects of variation
Recent generative models for 2D images achieve impressive visual results, but clearly lack the ability to perform 3D reasoning. This heavily restricts the degree of control over generated objects as well as the possible applications of such models. In this work, we leverage recent advances in differentiable rendering to design a framewo...More
PPT (Upload PPT)