3D Textured Shape Recovery with Learned Geometric Priors

TL;DR

This paper introduces a method for 3D textured shape recovery from partial scans by integrating learned geometric priors and a novel bounding box adaptation to improve robustness against occlusions and scale variations.

Contribution

It proposes a new approach that incorporates learned geometric priors and a completeness-aware bounding box adaptation for better 3D shape reconstruction from partial data.

Findings

Improved reconstruction accuracy on partial scans.

Enhanced robustness to occlusions and scale differences.

Effective integration of pose prediction with shape recovery.

Abstract

3D textured shape recovery from partial scans is crucial for many real-world applications. Existing approaches have demonstrated the efficacy of implicit function representation, but they suffer from partial inputs with severe occlusions and varying object types, which greatly hinders their application value in the real world. This technical report presents our approach to address these limitations by incorporating learned geometric priors. To this end, we generate a SMPL model from learned pose prediction and fuse it into the partial input to add prior knowledge of human bodies. We also propose a novel completeness-aware bounding box adaptation for handling different levels of scales and partialness of partial scans.