Learning Implicit Fields for Generative Shape Modeling

Code

• github.com/czq142857/implicit-decoder
  OfficialIn papertf★ 0
• github.com/ahnobari/Range-GAN
  tf★ 8
• github.com/czq142857/im-net-pytorch
  pytorch★ 0
• github.com/czq142857/im-net
  tf★ 0

Abstract

We advocate the use of implicit fields for learning generative models of shapes and introduce an implicit field decoder, called IM-NET, for shape generation, aimed at improving the visual quality of the generated shapes. An implicit field assigns a value to each point in 3D space, so that a shape can be extracted as an iso-surface. IM-NET is trained to perform this assignment by means of a binary classifier. Specifically, it takes a point coordinate, along with a feature vector encoding a shape, and outputs a value which indicates whether the point is outside the shape or not. By replacing conventional decoders by our implicit decoder for representation learning (via IM-AE) and shape generation (via IM-GAN), we demonstrate superior results for tasks such as generative shape modeling, interpolation, and single-view 3D reconstruction, particularly in terms of visual quality. Code and supplementary material are available at https://github.com/czq142857/implicit-decoder.

Tasks

Reproductions