Papers › PointNeXt: Revisiting PointNet++ with Improved Training and Scaling Strategies

PointNeXt: Revisiting PointNet++ with Improved Training and Scaling Strategies

9 Jun 2022arXiv:2206.04670archive 2025-07-28

Guocheng Qian, Yuchen Li, Houwen Peng, Jinjie Mai, Hasan Abed Al Kader Hammoud, Mohamed Elhoseiny, Bernard Ghanem

PointNet++ is one of the most influential neural architectures for point cloud understanding. Although the accuracy of PointNet++ has been largely surpassed by recent networks such as PointMLP and Point Transformer, we find that a large portion of the performance gain is due to improved training strategies, i.e. data augmentation and optimization techniques, and increased model sizes rather than architectural innovations. Thus, the full potential of PointNet++ has yet to be explored. In this work, we revisit the classical PointNet++ through a systematic study of model training and scaling strategies, and offer two major contributions. First, we propose a set of improved training strategies that significantly improve PointNet++ performance. For example, we show that, without any change in architecture, the overall accuracy (OA) of PointNet++ on ScanObjectNN object classification can be raised from 77.9% to 86.1%, even outperforming state-of-the-art PointMLP. Second, we introduce an inverted residual bottleneck design and separable MLPs into PointNet++ to enable efficient and effective model scaling and propose PointNeXt, the next version of PointNets. PointNeXt can be flexibly scaled up and outperforms state-of-the-art methods on both 3D classification and segmentation tasks. For classification, PointNeXt reaches an overall accuracy of 87.7 on ScanObjectNN, surpassing PointMLP by 2.3%, while being 10x faster in inference. For semantic segmentation, PointNeXt establishes a new state-of-the-art performance with 74.9% mean IoU on S3DIS (6-fold cross-validation), being superior to the recent Point Transformer. The code and models are available at https://github.com/guochengqian/pointnext.

PaperPDFCodeCode Syntology ranMethod as code

In Syntology View this paper on Syntology, its page in Syntology's graph. That page adds which of its functions are also present in other papers' code Syntology holds, with dates, and its dated run record as JSON and a README badge. It also lists the repositories linked to the paper, the abstract and the calls for agents.

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

For agents, on Syntology's MCP service (how to connect):

Code

Syntology 17 code samples harvested from 4 repositories linked to this paper: 10 ran (of which 0 constructed an object rather than computing a result; 4 with no instrument failure: 0 honoured, 0 violated, 4 with no contract checked; 6 where Syntology's instrument failed) · 7 unverified. Where Syntology's instrument failed: 6 where the contract check could not run; failures of Syntology's instrument, not of the code.

By repository: official repository: 3 samples from 1 repository, 3 ran (3 with no instrument failure; 0 where Syntology's instrument failed); community (archive-listed): 7 samples from 1 repository, 5 ran (1 with no instrument failure; 4 where Syntology's instrument failed); community: 7 samples from 2 repositories, 2 ran (0 with no instrument failure; 2 where Syntology's instrument failed). The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

No sample is linked to this paper from 1 of its repositories that Syntology harvested for other papers:

Levels written between 2026-09-22 and 2026-09-24 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed. 2 of the 10 samples that ran have no recorded date.

guochengqian/pointnext officialmentioned in papermentioned on GitHubpytorchMITSyntology: 3 samples, 3 ran (0 instrument-failed) · 0 unverified report
boyden/pointtransformerfl mentioned on GitHubpytorchSyntology: 7 samples, 5 ran (4 instrument-failed) · 2 unverified report
linhaojia13/pointmetabase mentioned on GitHubpytorchMITSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to CVF open access id Lin_Meta_Architecture_for_Point_Cloud_Analysis_CVPR_2023_paper (that paper's own run record: 10 ran (of which 0 constructed an object rather than computing a result; 7 with no instrument failure: 3 honoured, 1 violated, 3 with no contract checked; 3 where Syntology's instrument failed) · 1 unverified); nothing checks that they implement this paper's method. 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.

A paper named beside a repository with no sample linked to this paper is shown with that paper's own run record, not this paper's: “ran” means executed on a synthesized input, not that the code is correct, and “where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Papers are listed in arXiv-id order, at most three per repository.

Code Syntology ran Syntology

17 samples harvested: 10 ran (of which 0 constructed an object rather than computing a result; 4 with no instrument failure: 0 honoured, 0 violated, 4 with no contract checked; 6 where Syntology's instrument failed) · 7 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page has not changed since , the first build that kept a record date for it; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

6ran; the contract check could not run
4ran
7unverified

Licence: 0 of the 17 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 4 repositories linked to this paper, official or community; the list below is grouped by repository, each group headed by its name and what it is to the paper. “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. A line beginning “fixture:” under a sample that ran gives the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right. “Level recorded” gives the UTC date when Syntology's graph wrote the sample's verification level, not when the code executed; where the level was written before Syntology's graph kept that date, the row says “level recorded, date not recorded”. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself. “Our draft was wrong” and “the contract check could not run” are failures of Syntology's instrument, not of the code: the sample ran on Syntology's synthesized input, and then the drafted contract did not describe it, or the check itself raised an error (RAISES), lacked a dependency (DEP_MISSING) or ran out of time (TIMEOUT); hover the label for which. A metamorphic tier, where shown, is evidence of the output's shape, not that the sample implements the method.

Each sample ends with its code_sha256. A sample's code_sha256 is Syntology's identity for that exact code. It is printed in full (16 hexadecimal characters for every sample in this build), and get_code(code_sha256="…") takes it exactly as printed. An agent calls it on Syntology's MCP service (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 are grouped by repository: a repository marked official first, then by how directly the repository is tied to the paper, then in the archive's order; never by how many ran. 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 six 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; the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar); this paper's copy was not located, so its licence is not recorded; or this paper's copy is cleared, but the stored copy of the same code that Syntology serves is not. 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.

A function or class name links to its page when the name occurs in the harvested code of at least 5 papers; the name is a string, not an identity. A census tag appears when Syntology's census ran implementations with that name on one shared input per bucket: it gives how many were compared, from how many papers, and how many distinct outputs they produced, and names this sample's bucket when this sample was one of them.

guochengqian/pointnext official repository 3 samples, 3 ran (0 instrument-failed) · 0 unverified
parse_sha256_manifest pointnext_official/checkpoints.py ran MIT (permissive) · 1fd69c5312ba853f · report fixture: signature lines:list/1/str · returned dict · level recorded 2026-09-24
sha256_file in 21 papers pointnext_official/checkpoints.py ran MIT (permissive) · 7f2983739479f50d · report fixture: signature path:str/0/str, chunk_size:int/0/int · returned str · level recorded 2026-09-24
verify_sha256 pointnext_official/checkpoints.py ran MIT (permissive) · 9c821e95b597f78e · report fixture: signature path:str/0/str, expected_sha256:str/0/str · returned str · level recorded 2026-09-24
boyden/pointtransformerfl community (archive-listed) 7 samples, 5 ran (4 instrument-failed) · 2 unverified
WSConv models/point_transformer.py ran · metamorphic tier: invariant fingerprinted MIT (permissive) · 3b1ac6f82f83c1d9 · report fixture: signature x:Tensor/4/float · returned Tensor · level recorded 2026-09-22
index_points in 49 papers models/point_transformer.py ran; the contract check could not run (RAISES) MIT (permissive) · 6f566c4f5130b818 · report fixture: signature points:Tensor/3/float, idx:Tensor/2/int · returned Tensor · level recorded 2026-09-22
query_ball_point in 7 papers
census: 5 implementations from 5 papers, 1 distinct output on one shared input this sample: bucket 1 of 1, arg 1: rank 3, kind float, dtype float32 · arg 2: rank 3, kind float, dtype float32
models/point_transformer.py ran; the contract check could not run (RAISES) · metamorphic tier: invariant MIT (permissive) · 470b1e6c34a8774b · report fixture: signature radius:float/0/float, nsample:int/0/int, xyz:Tensor/3/float, new_xyz:Tensor/3/float · returned Tensor · level recorded 2026-09-22
sample_and_group in 10 papers models/point_transformer.py ran; the contract check could not run (RAISES) MIT (permissive) · 9809aa7254498ef8 · report fixture: signature npoint:int/0/int, radius:float/0/float, nsample:int/0/int, xyz:Tensor/3/float, points:Tensor/3/float, returnfps:bool/0/bool, knn:bool/0/bool, fps_idx:NoneType/0/? · returned tuple · level recorded 2026-09-24
square_distance in 89 papers
census: 11 implementations from 18 papers, 4 distinct outputs across 2 buckets, one shared input per bucket this sample: bucket 1 of 2, arg 1: rank 3, kind float, dtype float32 · arg 2: rank 3, kind float, dtype float32
models/point_transformer.py ran; the contract check could not run (RAISES) · metamorphic tier: invariant fingerprinted MIT (permissive) · cb46d51170c75900 · report fixture: signature src:Tensor/3/float, dst:Tensor/3/float · returned Tensor · level recorded 2026-09-22
PointNet2 models/point_transformer.py unverified MIT (permissive) · c0487a14ba8a5c22 · report
PointNetSetAbstraction in 7 papers models/point_transformer.py unverified MIT (permissive) · d0c7913713993782 · report
kentechx/pointnext community 6 samples, 2 ran (2 instrument-failed) · 4 unverified
cdist in 13 papers pointnext/pointnext.py ran; the contract check could not run (RAISES) · metamorphic tier: invariant fingerprinted MIT (permissive) · 23f479914e4bf067 · report fixture: signature x:Tensor/3/float, y:Tensor/3/float · returned Tensor · level recorded, date not recorded
downsample_fps pointnext/pointnext.py ran; the contract check could not run (RAISES) MIT (permissive) · 0ccbb20f77091ebe · report fixture: returned SampleResult · level recorded, date not recorded
ball_query pointnext/ops.py unverified MIT (permissive) · 58189ea84067b9ab · report
furthest_point_sample pointnext/ops.py unverified MIT (permissive) · 649b34ac4526091d · report
three_nn pointnext/ops.py unverified MIT (permissive) · 20c4b5f23e91db8f · report
yanx27/pointnet_pointnet2_pytorch community 1 sample, 0 ran · 1 unverified
test in 377 papers train_classification.py unverified MIT (permissive) · cce988c817575a1a · report

Method as code Syntology

What Syntology's graph records as this paper's claimed method, lifted into code: each assembly names its bricks (functions and classes), each with its own role and its own verification level. An assembly is not a check that the code implements the method correctly or reproduces the paper's results. Levels are each brick's own verification level in Syntology's graph (level 0: no recorded run), listed brick by brick and never combined into one level; “ran” means the brick executed on a synthesized input, not that its output is correct, and the instrument figure counts failures of Syntology's instrument, not of the code. Where a model read the code against the paper's claim, each assembly's label verdict is one model's judgement, kept in this page's JSON twin; it is not calibrated, so no card prints it.

1 assembly on this paper: 1 harvested, 0 generated.

harvested Assembly for PointNeXt (graph method: PointNet++). 9 bricks; individually on a fixture: 7 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 6 where Syntology's instrument failed) · 2 unverified. The assembly has not been run end-to-end.

  1. PointNet2 entry · level 0 · unverified · c0487a14ba8a5c22
  2. PointNetSetAbstraction model · level 0 · unverified · d0c7913713993782
  3. WSConv helper · level 1 · ran · 3b1ac6f82f83c1d9
  4. sample_and_group helper · level 1 · ran; the contract check could not run (RAISES) · 9809aa7254498ef8
  5. sample_and_group_all helper · level 1 · ran; the contract check could not run · 851a7e1fdd8b809a
  6. farthest_point_sample helper · level 1 · ran; the contract check could not run · 295c8424ac32f9cf
  7. index_points helper · level 1 · ran; the contract check could not run (RAISES) · 6f566c4f5130b818
  8. query_ball_point helper · level 1 · ran; the contract check could not run (RAISES) · 470b1e6c34a8774b
  9. square_distance helper · level 1 · ran; the contract check could not run (RAISES) · cb46d51170c75900

Each brick ends with its code_sha256, Syntology's identity for that exact code; click one to copy the get_code call that returns it. For agents, on Syntology's MCP service: assembly(arxiv_id="2206.04670") returns these assemblies with each brick's role and wiring (how to connect).

Tasks

3D Classification3D Part Segmentation3D Point Cloud Classification3D Semantic SegmentationData AugmentationSemantic SegmentationSupervised Only 3D Point Cloud Classification

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
3D Part Segmentation ShapeNet-Part PointNeXt Class Average IoU 85.2 #5 of 67 Archive leaderboard report
3D Part Segmentation ShapeNet-Part PointNeXt Instance Average IoU 87.1 #5 of 67 Archive leaderboard report
3D Point Cloud Classification ModelNet40 PointNeXt FLOPs 6.5G #28 of 111 Archive leaderboard report
3D Point Cloud Classification ModelNet40 PointNeXt Mean Accuracy 91.1 #28 of 111 Archive leaderboard report
3D Point Cloud Classification ModelNet40 PointNeXt Number of params 4.5M #28 of 111 Archive leaderboard report
3D Point Cloud Classification ModelNet40 PointNeXt Overall Accuracy 94.0 #28 of 111 Archive leaderboard report
3D Point Cloud Classification ScanObjectNN PointNeXt FLOPs 1.64G #40 of 77 Archive leaderboard report
3D Point Cloud Classification ScanObjectNN PointNeXt Mean Accuracy 86.8 #40 of 77 Archive leaderboard report
3D Point Cloud Classification ScanObjectNN PointNeXt Number of params 1.4M #40 of 77 Archive leaderboard report
3D Point Cloud Classification ScanObjectNN PointNeXt Overall Accuracy 88.2 #40 of 77 Archive leaderboard report
3D Semantic Segmentation OpenTrench3D PointNeXt-XL Model Size 41.5M #3 of 3 Archive leaderboard report
3D Semantic Segmentation OpenTrench3D PointNeXt-XL mAcc 79.7 #3 of 3 Archive leaderboard report
3D Semantic Segmentation OpenTrench3D PointNeXt-XL mIoU 70.6 #3 of 3 Archive leaderboard report
3D Semantic Segmentation S3DIS PointNext mIoU (6-Fold) 74.9 #3 of 6 Archive leaderboard report
3D Semantic Segmentation S3DIS PointNext mIoU (Area-5) 70.5 #3 of 6 Archive leaderboard report
Semantic Segmentation S3DIS PointNeXt-XL FLOPs 84.8G #12 of 54 Archive leaderboard report
Semantic Segmentation S3DIS PointNeXt-XL Mean IoU 74.9 #12 of 54 Archive leaderboard report
Semantic Segmentation S3DIS PointNeXt-XL Number of params 41.6M #12 of 54 Archive leaderboard report
Semantic Segmentation S3DIS PointNeXt-XL Params (M) 41.6 #12 of 54 Archive leaderboard report
Semantic Segmentation S3DIS PointNeXt-XL mAcc 83.0 #12 of 54 Archive leaderboard report
Semantic Segmentation S3DIS PointNeXt-XL oAcc 90.3 #12 of 54 Archive leaderboard report
Semantic Segmentation S3DIS PointNeXt-L FLOPs 15.2G #16 of 54 Archive leaderboard report
Semantic Segmentation S3DIS PointNeXt-L Mean IoU 73.9 #16 of 54 Archive leaderboard report
Semantic Segmentation S3DIS PointNeXt-L Number of params 7.1M #16 of 54 Archive leaderboard report
Semantic Segmentation S3DIS PointNeXt-L Params (M) 7.1 #16 of 54 Archive leaderboard report
Semantic Segmentation S3DIS PointNeXt-L mAcc 82.2 #16 of 54 Archive leaderboard report
Semantic Segmentation S3DIS PointNeXt-L oAcc 89.9 #16 of 54 Archive leaderboard report
Semantic Segmentation S3DIS Area5 PointNeXt Number of params 41.6M #29 of 61 Archive leaderboard report
Semantic Segmentation S3DIS Area5 PointNeXt mAcc 77.2 #29 of 61 Archive leaderboard report
Semantic Segmentation S3DIS Area5 PointNeXt mIoU 71.1 #29 of 61 Archive leaderboard report
Semantic Segmentation S3DIS Area5 PointNeXt oAcc 91.0 #29 of 61 Archive leaderboard report
Supervised Only 3D Point Cloud Classification ScanObjectNN PointNeXt GFLOPs 3.6 #6 of 12 Archive leaderboard report
Supervised Only 3D Point Cloud Classification ScanObjectNN PointNeXt Number of params (M) 1.4 #6 of 12 Archive leaderboard report
Supervised Only 3D Point Cloud Classification ScanObjectNN PointNeXt Overall Accuracy (PB_T50_RS) 87.8 #6 of 12 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 ConnectionSoftmaxTransformer

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