Papers › A-CNN: Annularly Convolutional Neural Networks on Point Clouds

A-CNN: Annularly Convolutional Neural Networks on Point Clouds

16 Apr 2019CVPR 2019 6arXiv:1904.08017archive 2025-07-28

Artem Komarichev, Zichun Zhong, Jing Hua

Analyzing the geometric and semantic properties of 3D point clouds through the deep networks is still challenging due to the irregularity and sparsity of samplings of their geometric structures. This paper presents a new method to define and compute convolution directly on 3D point clouds by the proposed annular convolution. This new convolution operator can better capture the local neighborhood geometry of each point by specifying the (regular and dilated) ring-shaped structures and directions in the computation. It can adapt to the geometric variability and scalability at the signal processing level. We apply it to the developed hierarchical neural networks for object classification, part segmentation, and semantic segmentation in large-scale scenes. The extensive experiments and comparisons demonstrate that our approach outperforms the state-of-the-art methods on a variety of standard benchmark datasets (e.g., ModelNet10, ModelNet40, ShapeNet-part, S3DIS, and ScanNet).

PaperPDFConference PDFCode

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

artemkomarichev/a-cnn officialmentioned on GitHubtf report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

3D Point Cloud ClassificationSegmentationSemantic Segmentation

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
3D Point Cloud Classification ModelNet40 A-CNN Overall Accuracy 92.6 #84 of 111 Archive leaderboard report
Semantic Segmentation S3DIS PointCNN Mean IoU 65.4 #35 of 54 Archive leaderboard report
Semantic Segmentation S3DIS PointCNN Number of params N/A #35 of 54 Archive leaderboard report
Semantic Segmentation S3DIS PointCNN oAcc 88.1 #35 of 54 Archive leaderboard report
Semantic Segmentation S3DIS A-CNN Mean IoU 62.9 #39 of 54 Archive leaderboard report
Semantic Segmentation S3DIS A-CNN Number of params N/A #39 of 54 Archive leaderboard report
Semantic Segmentation S3DIS A-CNN oAcc 87.3 #39 of 54 Archive leaderboard report
Semantic Segmentation S3DIS SPGraph Mean IoU 62.1 #41 of 54 Archive leaderboard report
Semantic Segmentation S3DIS SPGraph Number of params N/A #41 of 54 Archive leaderboard report
Semantic Segmentation S3DIS SPGraph oAcc 85.5 #41 of 54 Archive leaderboard report
Semantic Segmentation S3DIS 3P-RNN Mean IoU 56.3 #47 of 54 Archive leaderboard report
Semantic Segmentation S3DIS 3P-RNN Number of params N/A #47 of 54 Archive leaderboard report
Semantic Segmentation S3DIS 3P-RNN oAcc 86.9 #47 of 54 Archive leaderboard report
Semantic Segmentation S3DIS PointNet Mean IoU 47.6 #50 of 54 Archive leaderboard report
Semantic Segmentation S3DIS PointNet Number of params N/A #50 of 54 Archive leaderboard report
Semantic Segmentation S3DIS PointNet oAcc 78.5 #50 of 54 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

Convolution

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