Official PyTorch implementation of "Image-to-Lidar Self-Supervised Distillation for Autonomous Driving Data"

Fine-grained Image-to-LiDAR Contrastive Distillation with Visual Foundation Models

(AAAI2024) Point-PEFT: Parameter-Efficient Fine-Tuning for 3D Pre-trained Models

Code for our CVPR 2019 paper, HPLFlowNet: Hierarchical Permutohedral Lattice FlowNet for Scene Flow Estimation on Large-scale Point Clouds.

EFGHNet: A Versatile Image-to-Point Cloud Registration Network for Extreme Outdoor Environment

BEVContrast: Self-Supervision in BEV Space for Automotive Lidar Point Clouds - Official PyTorch implementation

Implementation of the paper "Multi-view PointNet for 3D Scene Understanding"

Self-supervised Learning of LiDAR 3D PointClouds via 2D-3D Neural Calibration

[TCSVT] CorrI2P: Deep Image-to-Point Cloud Registration via Dense CorrespondenceThe code of CorrI2P

[ICCV 2023] Take-A-Photo: 3D-to-2D Generative Pre-training of Point Cloud Models

Relational Priors Distillation

Baseline for Point Cloud Registration via Direct Superpoints Matching

Official repository of EP2P-Loc: End-to-End 3D Point to 2D Pixel Localization for Large-Scale Visual Localization (ICCV 2023)

Cross-Modal Information-Guided Network using Contrastive Learning for Point Cloud Registration (RAL 2023)

object-level 2d-3d registration demo code

Image-to-point cloud Registration Tool

DeepI2P: Image-to-Point Cloud Registration via Deep Classification. CVPR 2021

3D object detection

LIVW-Localization: A Multi-modal Information Fused Vehicle Localization method for Complex, Large-Scale and GNSS-Denied Environments.

PyTorch implementation of the paper: CMR-Agent: Learning a Cross-Modal Agent for Iterative Image-to-Point Cloud Registration (IROS 2024).

[NeurIPS'2023 Spotlight]: Differentiable Registration of Images and LiDAR Point Clouds with VoxelPoint-to-Pixel Matching

LIVW-Localization: A Multi-modal Information Fused Vehicle Localization Method for Complex, Large-Scale and GNSS-Denied Environments.

[IEEE RA-L 2024] CoFiI2P: Coarse-to-Fine Correspondences-Based Image-to-Point Cloud Registration

Code for "CMRNet: Camera to LiDAR-Map Registration" (ITSC 2019)

Attention-Enhanced Cross-modal Localization Between Spherical Images and Point Clouds

A fast and robust point cloud registration library