Papers › LCPFormer: Towards Effective 3D Point Cloud Analysis via Local Context Propagation in...
LCPFormer: Towards Effective 3D Point Cloud Analysis via Local Context Propagation in Transformers
Zhuoxu Huang, Zhiyou Zhao, Banghuai Li, Jungong Han
Transformer with its underlying attention mechanism and the ability to capture long-range dependencies makes it become a natural choice for unordered point cloud data. However, separated local regions from the general sampling architecture corrupt the structural information of the instances, and the inherent relationships between adjacent local regions lack exploration, while local structural information is crucial in a transformer-based 3D point cloud model. Therefore, in this paper, we propose a novel module named Local Context Propagation (LCP) to exploit the message passing between neighboring local regions and make their representations more informative and discriminative. More specifically, we use the overlap points of adjacent local regions (which statistically show to be prevalent) as intermediaries, then re-weight the features of these shared points from different local regions before passing them to the next layers. Inserting the LCP module between two transformer layers results in a significant improvement in network expressiveness. Finally, we design a flexible LCPFormer architecture equipped with the LCP module. The proposed method is applicable to different tasks and outperforms various transformer-based methods in benchmarks including 3D shape classification and dense prediction tasks such as 3D object detection and semantic segmentation. Code will be released for reproduction.
In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.
Code
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
Results from the paper archive 2025-07-28
| Task | Dataset | Model | Metric | Value | Rank at snapshot | Leaderboard | Report |
|---|---|---|---|---|---|---|---|
| 3D Object Detection | SUN-RGBD val | LCPFormer | mAP@0.25 | 63.2 | #16 of 32 | Archive leaderboard | report |
| 3D Object Detection | SUN-RGBD val | LCPFormer | mAP@0.5 | 46.2 | #16 of 32 | Archive leaderboard | report |
| 3D Point Cloud Classification | ModelNet40 | LCPFormer | Mean Accuracy | 90.7 | #54 of 111 | Archive leaderboard | report |
| 3D Point Cloud Classification | ModelNet40 | LCPFormer | Overall Accuracy | 93.6 | #54 of 111 | Archive leaderboard | report |
| 3D Semantic Segmentation | SensatUrban | LCPFormer | mIoU | 63.4 | #1 of 8 | Archive leaderboard | report |
| Semantic Segmentation | S3DIS Area5 | LCPFormer | Number of params | N/A | #33 of 61 | Archive leaderboard | report |
| Semantic Segmentation | S3DIS Area5 | LCPFormer | mAcc | 76.8 | #33 of 61 | Archive leaderboard | report |
| Semantic Segmentation | S3DIS Area5 | LCPFormer | mIoU | 70.2 | #33 of 61 | Archive leaderboard | report |
| Semantic Segmentation | S3DIS Area5 | LCPFormer | oAcc | 90.8 | #33 of 61 | 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.
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