Spectral Graphormer: Spectral Graph-based Transformer for Egocentric Two-Hand Reconstruction using Multi-View Color Images

We propose a novel transformer-based framework that reconstructs two high fidelity hands from multi-view RGB images. Unlike existing hand pose estimation methods, where one typically trains a deep network to regress hand model parameters from single RGB image, we consider a more challenging problem...

Full description

Saved in:
Bibliographic Details
Published inarXiv.org
Main Authors Tze Ho Elden Tse, Mueller, Franziska, Shen, Zhengyang, Tang, Danhang, Beeler, Thabo, Dou, Mingsong, Zhang, Yinda, Petrovic, Sasa, Hyung Jin Chang, Taylor, Jonathan, Doosti, Bardia
Format Paper
LanguageEnglish
Published Ithaca Cornell University Library, arXiv.org 21.08.2023
Subjects
Color imagery
Datasets
Optimization
Pose estimation
Reconstruction
Synthetic data
Transformers
Online AccessGet full text

Cover

More Information
Summary:We propose a novel transformer-based framework that reconstructs two high fidelity hands from multi-view RGB images. Unlike existing hand pose estimation methods, where one typically trains a deep network to regress hand model parameters from single RGB image, we consider a more challenging problem setting where we directly regress the absolute root poses of two-hands with extended forearm at high resolution from egocentric view. As existing datasets are either infeasible for egocentric viewpoints or lack background variations, we create a large-scale synthetic dataset with diverse scenarios and collect a real dataset from multi-calibrated camera setup to verify our proposed multi-view image feature fusion strategy. To make the reconstruction physically plausible, we propose two strategies: (i) a coarse-to-fine spectral graph convolution decoder to smoothen the meshes during upsampling and (ii) an optimisation-based refinement stage at inference to prevent self-penetrations. Through extensive quantitative and qualitative evaluations, we show that our framework is able to produce realistic two-hand reconstructions and demonstrate the generalisation of synthetic-trained models to real data, as well as real-time AR/VR applications.
ISSN:2331-8422