This document summarizes a paper titled "DeepI2P: Image-to-Point Cloud Registration via Deep Classification". The paper proposes a method for estimating the camera pose within a point cloud map using a deep learning model. The model first classifies whether points in the point cloud fall within the camera's frustum or image grid. It then performs pose optimization to estimate the camera pose by minimizing the projection error of inlier points onto the image. The method achieves more accurate camera pose estimation compared to existing techniques based on feature matching or depth estimation. It provides a new approach for camera localization using point cloud maps without requiring cross-modal feature learning.
4. 紹介する論文
4
DeepI2P: Image-to-Point Cloud Registration via Deep
Classification
Jiaxin Li (Bytedance), Gim Hee Lee (National University of
Singapore)
選んだ理由:
個人的に興味のあるテーマ
8. Related Work: 2D3D-MatchNet
画像はSIFT、点群はISSによってキーポイントを抽出し、キー
ポイント間のマッチングを行うための特徴量をTriplet Lossを
用いて学習
Feng, M., Hu, S.,Ang, M., & Lee, G. H. (2019). 2D3D-MatchNet: Learning to Match Keypoints Across 2D Image and 3D Point Cloud.
International Conference on Robotics and Automation .
9. Related Work: 2D-3D Line Correspondences
画像と点群上の直線をマッチさせることで自己位置推定
Visual SLAMによるTrackingが前提条件
Yu, H., Zhen,W.,Yang,W., Zhang, J., & Scherer, S. (2020). Monocular Camera Localization in Prior LiDAR Maps with 2D-3D Line
Correspondences. IEEE International Conference on Intelligent Robots and Systems