Papers › Point Transformer V3: Simpler, Faster, Stronger

Point Transformer V3: Simpler, Faster, Stronger

15 Dec 2023arXiv:2312.10035archive 2025-07-28

Xiaoyang Wu, Li Jiang, Peng-Shuai Wang, Zhijian Liu, Xihui Liu, Yu Qiao, Wanli Ouyang, Tong He, Hengshuang Zhao

This paper is not motivated to seek innovation within the attention mechanism. Instead, it focuses on overcoming the existing trade-offs between accuracy and efficiency within the context of point cloud processing, leveraging the power of scale. Drawing inspiration from recent advances in 3D large-scale representation learning, we recognize that model performance is more influenced by scale than by intricate design. Therefore, we present Point Transformer V3 (PTv3), which prioritizes simplicity and efficiency over the accuracy of certain mechanisms that are minor to the overall performance after scaling, such as replacing the precise neighbor search by KNN with an efficient serialized neighbor mapping of point clouds organized with specific patterns. This principle enables significant scaling, expanding the receptive field from 16 to 1024 points while remaining efficient (a 3x increase in processing speed and a 10x improvement in memory efficiency compared with its predecessor, PTv2). PTv3 attains state-of-the-art results on over 20 downstream tasks that span both indoor and outdoor scenarios. Further enhanced with multi-dataset joint training, PTv3 pushes these results to a higher level.

PaperPDFCodeCode Syntology ran

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

For agents, Syntology's MCP tool lists every function and class Syntology harvested from this paper and whether it ran (how to connect): get_harvested_code_for_paper(arxiv_id="2312.10035")

Code

Syntology Ran 7 of 11 code samples harvested from 1 repository linked to this paper; 4 have no recorded run. Of those that ran: 7 ran with no contract checked.

By repository: official repository: 11 samples from 1 repository, 7 ran. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

Pointcept/Pointcept officialmentioned in papermentioned on GitHubpytorchMIT report
pointcept/pointtransformerv3 officialmentioned in papermentioned on GitHubpytorchMIT report
facebookresearch/SparseConvNet mentioned on GitHubpytorchNOASSERTION 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

11 samples harvested; 7 ran; 0 honoured the contract we drafted; 4 have no recorded run. Read from Syntology's graph 2026-09-24; that is when this build read the record, not when the samples ran.

7ran
4unverified

Licence: 0 of the 11 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from pointcept/pointtransformerv3. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself; “our draft was wrong” and “fixture could not drive it” are failures of Syntology's instrument, not of the code.

Each sample ends with its code_sha256, Syntology's identity for that exact code. An agent fetches the stored sample with Syntology's MCP tool get_code(code_sha256="…") (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples from a repository marked official are listed first. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of four reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; or the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar). Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

batch2offset pointcept/pointtransformerv3/model.py official repository ran MIT (permissive) · c02237104f832f3d · report
gray2binary pointcept/pointtransformerv3/serialization/hilbert.py official repository ran MIT (permissive) · f3ce15083e63461d · report
key2xyz pointcept/pointtransformerv3/serialization/z_order.py official repository ran MIT (permissive) · 4c4e7aa66bbabbff · report
offset2batch pointcept/pointtransformerv3/model.py official repository ran MIT (permissive) · 784720b30e505460 · report
offset2bincount pointcept/pointtransformerv3/model.py official repository ran MIT (permissive) · c17c79b2e37251f8 · report
right_shift pointcept/pointtransformerv3/serialization/hilbert.py official repository ran fingerprinted MIT (permissive) · 7af58291b04f4410 · report
xyz2key pointcept/pointtransformerv3/serialization/z_order.py official repository ran MIT (permissive) · 9a000022d903a2f0 · report
binary2gray pointcept/pointtransformerv3/serialization/hilbert.py official repository unverified MIT (permissive) · 48079b763eddc476 · report
decode pointcept/pointtransformerv3/serialization/default.py official repository unverified MIT (permissive) · 89ba8581eca2a6a5 · report
encode pointcept/pointtransformerv3/serialization/default.py official repository unverified MIT (permissive) · ef0574663c9ae24c · report
z_order_encode pointcept/pointtransformerv3/serialization/default.py official repository unverified MIT (permissive) · 4f4ada46d5688580 · report

Tasks

3D Semantic SegmentationLIDAR Semantic SegmentationRepresentation LearningSemantic Segmentation

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
3D Semantic Segmentation ScanNet++ PTv3 Top-1 IoU 0.488 #3 of 8 Archive leaderboard report
3D Semantic Segmentation ScanNet++ PTv3 Top-3 IoU 0.725 #3 of 8 Archive leaderboard report
3D Semantic Segmentation ScanNet200 PTv3 + PPT test mIoU 39.3 #7 of 16 Archive leaderboard report
3D Semantic Segmentation ScanNet200 PTv3 + PPT val mIoU 36.0 #7 of 16 Archive leaderboard report
3D Semantic Segmentation SemanticKITTI PPT+PTv3 test mIoU 75.5% #2 of 45 Archive leaderboard report
3D Semantic Segmentation SemanticKITTI PPT+PTv3 val mIoU 72.3% #2 of 45 Archive leaderboard report
LIDAR Semantic Segmentation nuScenes PTv3 + PPT test mIoU 0.830 #2 of 36 Archive leaderboard report
LIDAR Semantic Segmentation nuScenes PTv3 + PPT val mIoU 0.812 #2 of 36 Archive leaderboard report
Semantic Segmentation S3DIS PTv3 + PPT Mean IoU 80.8 #2 of 54 Archive leaderboard report
Semantic Segmentation S3DIS PTv3 + PPT Number of params 24.1M #2 of 54 Archive leaderboard report
Semantic Segmentation S3DIS PTv3 + PPT mAcc 87.7 #2 of 54 Archive leaderboard report
Semantic Segmentation S3DIS PTv3 + PPT oAcc 92.6 #2 of 54 Archive leaderboard report
Semantic Segmentation S3DIS Area5 PTv3 + PPT mAcc 80.1 #4 of 61 Archive leaderboard report
Semantic Segmentation S3DIS Area5 PTv3 + PPT mIoU 74.7 #4 of 61 Archive leaderboard report
Semantic Segmentation S3DIS Area5 PTv3 + PPT oAcc 92.0 #4 of 61 Archive leaderboard report
Semantic Segmentation ScanNet PTv3 + PPT test mIoU 79.4 #4 of 45 Archive leaderboard report
Semantic Segmentation ScanNet PTv3 + PPT val mIoU 78.6 #4 of 45 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

Absolute Position EncodingsAdamAttentionBPEDense ConnectionsDropoutLabel SmoothingLayer NormalizationLinear LayerMulti-Head AttentionPosition-Wise Feed-Forward LayerResidual ConnectionSPEEDSoftmaxTransformer

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