2D3D-MATR: 2D-3D Matching Transformer for Detection-free Registration between Images and Point Clouds

TL;DR

The paper introduces 2D3D-MATR, a detection-free transformer-based method for accurate registration between images and point clouds, overcoming cross-modality challenges with a coarse-to-fine matching approach.

Contribution

It proposes a novel detection-free registration method using transformers and multi-scale patch matching, improving robustness and accuracy over existing methods.

Findings

Outperforms P2-Net by 20% in inlier ratio

Achieves over 10% higher registration recall

Demonstrates robustness on public benchmarks

Abstract

The commonly adopted detect-then-match approach to registration finds difficulties in the cross-modality cases due to the incompatible keypoint detection and inconsistent feature description. We propose, 2D3D-MATR, a detection-free method for accurate and robust registration between images and point clouds. Our method adopts a coarse-to-fine pipeline where it first computes coarse correspondences between downsampled patches of the input image and the point cloud and then extends them to form dense correspondences between pixels and points within the patch region. The coarse-level patch matching is based on transformer which jointly learns global contextual constraints with self-attention and cross-modality correlations with cross-attention. To resolve the scale ambiguity in patch matching, we construct a multi-scale pyramid for each image patch and learn to find for each point patch the best matching image patch at a proper resolution level. Extensive experiments on two public benchmarks demonstrate that 2D3D-MATR outperforms the previous state-of-the-art P2-Net by around $20$ percentage points on inlier ratio and over $10$ points on registration recall. Our code and models are available at https://github.com/minhaolee/2D3DMATR.