TransPose: 6D Object Pose Estimation with Geometry-Aware Transformer

TL;DR

TransPose introduces a geometry-aware Transformer-based framework for 6D object pose estimation from RGB data, effectively handling occlusion and illumination variations by integrating local and global geometric features.

Contribution

The paper presents a novel Transformer encoder with a geometry-aware module that enhances point cloud feature learning for improved 6D pose estimation.

Findings

Achieves competitive results on benchmark datasets.

Effectively handles occlusion and illumination changes.

Integrates local geometry with global context via Transformer.

Abstract

Estimating the 6D object pose is an essential task in many applications. Due to the lack of depth information, existing RGB-based methods are sensitive to occlusion and illumination changes. How to extract and utilize the geometry features in depth information is crucial to achieve accurate predictions. To this end, we propose TransPose, a novel 6D pose framework that exploits Transformer Encoder with geometry-aware module to develop better learning of point cloud feature representations. Specifically, we first uniformly sample point cloud and extract local geometry features with the designed local feature extractor base on graph convolution network. To improve robustness to occlusion, we adopt Transformer to perform the exchange of global information, making each local feature contains global information. Finally, we introduce geometry-aware module in Transformer Encoder, which to form an effective constrain for point cloud feature learning and makes the global information exchange more tightly coupled with point cloud tasks. Extensive experiments indicate the effectiveness of TransPose, our pose estimation pipeline achieves competitive results on three benchmark datasets.