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

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

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

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

[ICML 2022] Benchmarking and Analyzing Point Cloud Classification under Corruptions https://arxiv.org/abs/2202.03377

[IROS 2020] Indoor Scene Recognition in 3D

Official code of UFORecon (CVPR 2024)

PyTorchGeoNodes is a PyTorch module for differentiable shape programs / procedural models in forms of graphs. It can automatically translate Blender geometry node models into PyTorch code. Originally, it was designed to simplify the integration of procedural shape programs into machine learning pipelines for 3D scene understanding.

Pytorch code for "DPFM: Deep Partial Functional Maps" - 3DV 2021 (Oral)