# Angle-I2P: Angle-Consistent-Aware Hierarchical Attention for Cross-Modality Outlier Rejection

arXiv: https://arxiv.org/abs/2605.04541 · HTML: https://mlsanity.com/paper/2605.04541/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 16 of its 31 references that are in the map.
- First posted on arXiv 6 May 2026.

## Builds on (most cited first, 16 of 16)

- [Depth Anything V2](https://mlsanity.com/paper/2406.09414/) (2024)
- [TEASER: Fast and Certifiable Point Cloud Registration](https://mlsanity.com/paper/2001.07715/) (2020)
- [MASt3R-SLAM: Real-Time Dense SLAM with 3D Reconstruction Priors](https://mlsanity.com/paper/2412.12392/) (2024)
- [Graph-Cut RANSAC](https://mlsanity.com/paper/1706.00984/) (2017)
- [3D Registration with Maximal Cliques](https://mlsanity.com/paper/2305.10854/) (2023)
- [2D3D-MatchNet: Learning to Match Keypoints Across 2D Image and 3D Point Cloud](https://mlsanity.com/paper/1904.09742/) (2019)
- [DeepI2P: Image-to-Point Cloud Registration via Deep Classification](https://mlsanity.com/paper/2104.03501/) (2021)
- [Is Geometry Enough for Matching in Visual Localization?](https://mlsanity.com/paper/2203.12979/) (2022)
- [P2-Net: Joint Description and Detection of Local Features for Pixel and Point Matching](https://mlsanity.com/paper/2103.01055/) (2021)
- [FreeReg: Image-to-Point Cloud Registration Leveraging Pretrained Diffusion Models and Monocular Depth Estimators](https://mlsanity.com/paper/2310.03420/) (2023)
- [LCD: Learned Cross-Domain Descriptors for 2D-3D Matching](https://mlsanity.com/paper/1911.09326/) (2019)
- [2D3D-MATR: 2D-3D Matching Transformer for Detection-free Registration between Images and Point Clouds](https://mlsanity.com/paper/2308.05667/) (2023)
- [An Economic Framework for 6-DoF Grasp Detection](https://mlsanity.com/paper/2407.08366/) (2024)
- [FastMAC: Stochastic Spectral Sampling of Correspondence Graph](https://mlsanity.com/paper/2403.08770/) (2024)
- [Deep Graph-based Spatial Consistency for Robust Non-rigid Point Cloud Registration](https://mlsanity.com/paper/2303.09950/) (2023)
- [HyperGCT: A Dynamic Hyper-GNN-Learned Geometric Constraint for 3D Registration](https://mlsanity.com/paper/2503.02195/) (2025)

## Abstract (arXiv)

Image-to-point-cloud registration (I2P) is a fundamental task in robotic applications such as manipulation,grasping, and localization. Existing deep learning-based I2P methods seek to align image and point cloud features in a learned representation space to establish correspondences, and have achieved promising results. However, when the inlier ratio of the initial matching pairs is low, conventional Perspective-n-Points (PnP) methods may struggle to achieve accurate results. To address this limitation, we propose Angle-I2P, an outlier rejection network that leverages angle-consistent geometric constraints and hierarchical attention. First, we design a scale-invariant, crossmodality geometric constraint based on angular consistency. This explicit geometric constraint guides the model in distinguishing inliers from outliers. Furthermore, we propose a global-tolocal hierarchical attention mechanism that effectively filters out geometrically inconsistent matches under rigid transformation, thereby improving the Inlier Ratio (IR) and Registration Recall (RR). Experimental results demonstrate that our method achieves state-of-the-art performance on the 7Scenes, RGBD Scenes V2, and a self-collected dataset, with consistent improvements across all benchmarks.