Datasets › 3DMatch

3DMatch

Introduced by Andy Zeng et al. in 3DMatch: Learning Local Geometric Descriptors from RGB-D Reconstructions archive 2025-07-28

The 3DMATCH benchmark evaluates how well descriptors (both 2D and 3D) can establish correspondences between RGB-D frames of different views. The dataset contains 2D RGB-D patches and 3D patches (local TDF voxel grid volumes) of wide-baselined correspondences.

The pixel size of each 2D patch is determined by the projection of the 0.3m3 local 3D patch around the interest point onto the image plane.

Source: 3DMatch: Learning Local Geometric Descriptors from RGB-D Reconstructions

Homepage Benchmarks Papers JSON

Benchmarks archive 2025-07-28

All 3 leaderboards whose dataset resolves to this page shown (sort by any header). "First row" is the archive's own first row at snapshot, in the archive's row order; nothing here re-ranks and metric direction is not asserted.

		First row (archive order)	Paper	Code
Point Cloud Registration	3DMatch Benchmark	IMFNet Feature Matching Recall 98.6	IMFNet: Interpretable Multimodal Fusion for Point Cloud...	XiaoshuiHuang/IMFNet	15	Compare
3D Feature Matching	3DMatch Benchmark	FCGF Average Recall 0.9578	Fully Convolutional Geometric Features	chrischoy/FCGF	1	Compare
Low-Light Image Enhancement	3DMatch Benchmark	nnh mAP (@0.1, Through-wall) 224	Attention Guided Low-light Image Enhancement with a...	Lvfeifan/MBLLEN +2	1	Compare

Papers archive 2025-07-28

15 shown of 15 papers with a leaderboard row on this dataset's benchmarks, newest first. The archive's own "papers using this dataset" list was never published, so this is the benchmark-backed subset; the archive's count for this dataset is 175. The Syntology column is from Syntology's graph (read 2026-09-24), stated per sample; it is not part of any archive number.

			Date	Samples run Syntology
Points to Patches: Enabling the Use of Self-Attention for 3D Shape Recognition	1	1	8 Apr 2022	not harvested
IMFNet: Interpretable Multimodal Fusion for Point Cloud Registration	1	1	18 Nov 2021	not harvested
Learning general and distinctive 3D local deep descriptors for point cloud registration	1	1	21 May 2021	not harvested
3D Point Cloud Registration with Multi-Scale Architecture and Unsupervised Transfer Learning	1	1	26 Mar 2021	not harvested
SpinNet: Learning a General Surface Descriptor for 3D Point Cloud Registration	1	1	24 Nov 2020	ran 0 of 1 samples (1 unverified)
Distinctive 3D local deep descriptors	2	1	1 Sep 2020	not harvested
End-to-End Learning Local Multi-view Descriptors for 3D Point Clouds	1	1	12 Mar 2020	ran 1 of 13 samples (12 unverified)
D3Feat: Joint Learning of Dense Detection and Description of 3D Local Features	2	2	6 Mar 2020	ran 0 of 8 samples (8 unverified)
Fully Convolutional Geometric Features	1	2	27 Oct 2019	not harvested
Attention Guided Low-light Image Enhancement with a Large Scale Low-light Simulation Dataset	3	1	2 Aug 2019	not harvested
The Perfect Match: 3D Point Cloud Matching with Smoothed Densities	2	1	16 Nov 2018	ran 1 of 8 samples (7 unverified)
PPF-FoldNet: Unsupervised Learning of Rotation Invariant 3D Local Descriptors	2	1	30 Aug 2018	ran 1 of 2 samples (1 unverified; 2 pointer-only for licence)
PPFNet: Global Context Aware Local Features for Robust 3D Point Matching	1	1	7 Feb 2018	not harvested
3DMatch: Learning Local Geometric Descriptors from RGB-D Reconstructions	2	1	27 Mar 2016	not harvested
Fast Point Feature Histograms (FPFH) for 3D Registration	1	1	12 May 2009	not harvested

Dataset loaders archive 2025-07-28

No loader listed in the archive.

Tasks archive 2025-07-28

Point Cloud Registration Low-Light Image Enhancement 3D Feature Matching

License archive 2025-07-28

Various

Modalities archive 2025-07-28

No modality tagged.

Languages archive 2025-07-28

No language tagged.

Variants archive 2025-07-28

3DMatch Benchmark
3DMatch

2 variant names, as the archive lists them.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections