1st Place Solution to the 8th HANDS Workshop Challenge -- ARCTIC Track: 3DGS-based Bimanual Category-agnostic Interaction Reconstruction

TL;DR

This paper presents a top-performing method for reconstructing 3D models of hands and objects from monocular videos in bimanual interactions, using novel loss functions and 3D Gaussian Splatting, achieving state-of-the-art results.

Contribution

The paper introduces a new approach combining mask and contact losses with 3D Gaussian Splatting for category-agnostic 3D interaction reconstruction from monocular videos.

Findings

Achieved a CD_h score of 38.69 on the ARCTIC test set.

Introduced mask loss and 3D contact loss to handle occlusion and contact.

Demonstrated effectiveness of 3D Gaussian Splatting in this task.

Abstract

This report describes our 1st place solution to the 8th HANDS workshop challenge (ARCTIC track) in conjunction with ECCV 2024. In this challenge, we address the task of bimanual category-agnostic hand-object interaction reconstruction, which aims to generate 3D reconstructions of both hands and the object from a monocular video, without relying on predefined templates. This task is particularly challenging due to the significant occlusion and dynamic contact between the hands and the object during bimanual manipulation. We worked to resolve these issues by introducing a mask loss and a 3D contact loss, respectively. Moreover, we applied 3D Gaussian Splatting (3DGS) to this task. As a result, our method achieved a value of 38.69 in the main metric, CD$_h$, on the ARCTIC test set.