Neural 3D Mesh Renderer
This work addresses the challenge of modeling 3D meshes from 2D images for computer vision and graphics applications, representing a novel method for a known bottleneck.
The paper tackled the problem of integrating polygon mesh rendering into neural networks by proposing an approximate gradient for rasterization, enabling single-image 3D mesh reconstruction that outperforms existing voxel-based approaches and allowing gradient-based 3D mesh editing operations like 2D-to-3D style transfer.
For modeling the 3D world behind 2D images, which 3D representation is most appropriate? A polygon mesh is a promising candidate for its compactness and geometric properties. However, it is not straightforward to model a polygon mesh from 2D images using neural networks because the conversion from a mesh to an image, or rendering, involves a discrete operation called rasterization, which prevents back-propagation. Therefore, in this work, we propose an approximate gradient for rasterization that enables the integration of rendering into neural networks. Using this renderer, we perform single-image 3D mesh reconstruction with silhouette image supervision and our system outperforms the existing voxel-based approach. Additionally, we perform gradient-based 3D mesh editing operations, such as 2D-to-3D style transfer and 3D DeepDream, with 2D supervision for the first time. These applications demonstrate the potential of the integration of a mesh renderer into neural networks and the effectiveness of our proposed renderer.