3dvision

[ICCV 2021 Oral] PoinTr: Diverse Point Cloud Completion with Geometry-Aware Transformers

[CVPR 2021, Oral] PREDATOR: Registration of 3D Point Clouds with Low Overlap.

[ICCV 2021 Oral] NerfingMVS: Guided Optimization of Neural Radiance Fields for Indoor Multi-view Stereo

3DV 2021: Synergy between 3DMM and 3D Landmarks for Accurate 3D Facial Geometry

[3DV 2025, Oral] LoopSplat: Loop Closure by Registering 3D Gaussian Splats

Mix3D: Out-of-Context Data Augmentation for 3D Scenes (3DV 2021 Oral)

a comprehensive and critical synthesis of the emerging role of GenAI across the full autonomous driving stack

Official repository for the ShapeFormer Project

(TPAMI 2023) Snowflake Point Deconvolution for Point Cloud Completion and Generation with Skip-Transformer

Grid-GCN for Fast and Scalable Point Cloud Learning

Generative 3D Part Assembly via Dynamic Graph Learning, NeurIPS 2020

Benchmarking and Analyzing Point Cloud Perception Robustness under Corruptions

Official code of VolRecon (CVPR 2023)

[CVPR 24] MaskClustering: View Consensus based Mask Graph Clustering for Open-Vocabulary 3D Instance Segmentation

[ICCV'21] PlaneTR: Structure-Guided Transformers for 3D Plane Recovery

Configurable point cloud registration pipeline.

A curated list of awesome Single-view 3D Object Reconstruction papers & resources

RANSAC segmentation-based ground removal for 3D LiDAR point clouds.

Using adpative feature grid to reduce the memory of NICE-SLAM

[CVPR 2021] PV-RAFT: Point-Voxel Correlation Fields for Scene Flow Estimation of Point Clouds