TSCom-Net: Coarse-to-Fine 3D Textured Shape Completion Network

TL;DR

This paper introduces TSCom-Net, a neural network architecture that reconstructs full 3D human body shapes and high-resolution textures from partial scans, using a coarse-to-fine approach with implicit learning and texture inpainting.

Contribution

The paper presents a novel two-stage neural network for 3D body shape and texture completion, combining implicit shape reconstruction with high-resolution texture inpainting.

Findings

Achieves competitive results on 3DBodyTex.V2 dataset.

Ranks second in the SHARP 2022 challenge.

Generalizes well to various partial shapes.

Abstract

Reconstructing 3D human body shapes from 3D partial textured scans remains a fundamental task for many computer vision and graphics applications -- e.g., body animation, and virtual dressing. We propose a new neural network architecture for 3D body shape and high-resolution texture completion -- BCom-Net -- that can reconstruct the full geometry from mid-level to high-level partial input scans. We decompose the overall reconstruction task into two stages - first, a joint implicit learning network (SCom-Net and TCom-Net) that takes a voxelized scan and its occupancy grid as input to reconstruct the full body shape and predict vertex textures. Second, a high-resolution texture completion network, that utilizes the predicted coarse vertex textures to inpaint the missing parts of the partial 'texture atlas'. A thorough experimental evaluation on 3DBodyTex.V2 dataset shows that our method achieves competitive results with respect to the state-of-the-art while generalizing to different types and levels of partial shapes. The proposed method has also ranked second in the track1 of SHApe Recovery from Partial textured 3D scans (SHARP [38,1]) 2022 challenge1.