Papers › Back to the Feature: Classical 3D Features are (Almost) All You Need for 3D Anomaly Detection

Back to the Feature: Classical 3D Features are (Almost) All You Need for 3D Anomaly Detection

10 Mar 2022arXiv:2203.05550archive 2025-07-28

Eliahu Horwitz, Yedid Hoshen

Despite significant advances in image anomaly detection and segmentation, few methods use 3D information. We utilize a recently introduced 3D anomaly detection dataset to evaluate whether or not using 3D information is a lost opportunity. First, we present a surprising finding: standard color-only methods outperform all current methods that are explicitly designed to exploit 3D information. This is counter-intuitive as even a simple inspection of the dataset shows that color-only methods are insufficient for images containing geometric anomalies. This motivates the question: how can anomaly detection methods effectively use 3D information? We investigate a range of shape representations including hand-crafted and deep-learning-based; we demonstrate that rotation invariance plays the leading role in the performance. We uncover a simple 3D-only method that beats all recent approaches while not using deep learning, external pre-training datasets, or color information. As the 3D-only method cannot detect color and texture anomalies, we combine it with color-based features, significantly outperforming previous state-of-the-art. Our method, dubbed BTF (Back to the Feature) achieves pixel-wise ROCAUC: 99.3% and PRO: 96.4% on MVTec 3D-AD.

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="2203.05550")

Code

Syntology Ran 1 of 9 code samples harvested from 1 repository linked to this paper; 8 have no recorded run. Of those that ran: 1 ran · fixture could not drive it.

By repository: official repository: 8 samples from 1 repository, 0 ran; 1 identical to code first harvested elsewhere. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

eliahuhorwitz/3D-ADS officialmentioned on GitHubpytorchMIT 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

9 samples harvested; 1 ran; 0 honoured the contract we drafted; 8 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.

1ran · fixture could not drive it
8unverified

Licence: 1 of the 9 samples is 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 eliahuhorwitz/3D-ADS. Some samples are identical code Syntology first harvested from another repository; for those, this paper's copy is not located and its licence is not recorded. “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.

collect_anomaly_scores eliahuhorwitz/3D-ADS/utils/au_pro_util.py official repository unverified MIT (permissive) · 18c9bfa7e94951e9 · report
compute_pro eliahuhorwitz/3D-ADS/utils/au_pro_util.py official repository unverified MIT (permissive) · 49b5692178958afe · report
get_sift_bin_ksize_stride_pad eliahuhorwitz/3D-ADS/utils/DenseSIFTDescriptor.py official repository unverified MIT (permissive) · 501cb82dcd37b5e7 · report
get_sift_pooling_kernel eliahuhorwitz/3D-ADS/utils/DenseSIFTDescriptor.py official repository unverified MIT (permissive) · 21dbe36d5d11746a · report
organized_pc_to_unorganized_pc eliahuhorwitz/3D-ADS/utils/mvtec3d_util.py official repository unverified MIT (permissive) · 65012063890a92f8 · report
read_tiff_organized_pc eliahuhorwitz/3D-ADS/utils/mvtec3d_util.py official repository unverified MIT (permissive) · 45b5420cddbb1bde · report
resize_organized_pc eliahuhorwitz/3D-ADS/utils/mvtec3d_util.py official repository unverified MIT (permissive) · f4f7ac4f1e370dd3 · report
trapezoid eliahuhorwitz/3D-ADS/utils/au_pro_util.py official repository unverified MIT (permissive) · 9d72c1f1f0d6dea4 · report
get_edges_of_pc identical code first harvested elsewhere ran · fixture could not drive it fingerprinted licence of this copy not recorded · f776b0618055e2f4 · report

Tasks

3D Anomaly Detection3D Anomaly Detection and SegmentationAllAnomaly DetectionDepth Anomaly Detection and SegmentationRGB+3D Anomaly Detection and SegmentationRGB+Depth Anomaly Detection and Segmentation

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
3D Anomaly Detection Anomaly-ShapeNet10 BTF (FPFH) O-AUROC 0.632 #3 of 7 Archive leaderboard report
3D Anomaly Detection Anomaly-ShapeNet10 BTF (FPFH) P-AUROC 0.790 #3 of 7 Archive leaderboard report
3D Anomaly Detection Anomaly-ShapeNet10 BTF (Raw) O-AUROC 0.500 #7 of 7 Archive leaderboard report
3D Anomaly Detection Anomaly-ShapeNet10 BTF (Raw) P-AUROC 0.515 #7 of 7 Archive leaderboard report
3D Anomaly Detection Real 3D-AD BTF (Raw) Mean Performance of P. and O. 0.6785 #13 of 19 Archive leaderboard report
3D Anomaly Detection Real 3D-AD BTF (Raw) Object AUROC 0.635 #13 of 19 Archive leaderboard report
3D Anomaly Detection Real 3D-AD BTF (Raw) Point AUROC 0.722 #13 of 19 Archive leaderboard report
3D Anomaly Detection Real 3D-AD BTF (FPFH) Mean Performance of P. and O. 0.5845 #19 of 19 Archive leaderboard report
3D Anomaly Detection Real 3D-AD BTF (FPFH) Object AUROC 0.603 #19 of 19 Archive leaderboard report
3D Anomaly Detection Real 3D-AD BTF (FPFH) Point AUROC 0.566 #19 of 19 Archive leaderboard report
3D Anomaly Detection and Segmentation MVTEC 3D-AD Back to the Feature: Classical 3D Features are (Almost) All You Need for 3D Anomaly Detection (FPFH) Detection AUROC 0.782 #7 of 11 Archive leaderboard report
3D Anomaly Detection and Segmentation MVTEC 3D-AD Back to the Feature: Classical 3D Features are (Almost) All You Need for 3D Anomaly Detection (FPFH) Segmentation AUROC 0.978 #7 of 11 Archive leaderboard report
Depth Anomaly Detection and Segmentation MVTEC 3D-AD Back to the Feature: Classical 3D Features are (Almost) All You Need for 3D Anomaly Detection (SIFT) Detection AUROC 0.727 #5 of 13 Archive leaderboard report
Depth Anomaly Detection and Segmentation MVTEC 3D-AD Back to the Feature: Classical 3D Features are (Almost) All You Need for 3D Anomaly Detection (SIFT) Segmentation AUPRO 0.910 #5 of 13 Archive leaderboard report
Depth Anomaly Detection and Segmentation MVTEC 3D-AD Back to the Feature: Classical 3D Features are (Almost) All You Need for 3D Anomaly Detection (SIFT) Segmentation AUROC 0.974 #5 of 13 Archive leaderboard report
Depth Anomaly Detection and Segmentation MVTEC 3D-AD Back to the Feature: Classical 3D Features are (Almost) All You Need for 3D Anomaly Detection (HoG) Detection AUROC 0.559 #7 of 13 Archive leaderboard report
Depth Anomaly Detection and Segmentation MVTEC 3D-AD Back to the Feature: Classical 3D Features are (Almost) All You Need for 3D Anomaly Detection (HoG) Segmentation AUPRO 0.771 #7 of 13 Archive leaderboard report
Depth Anomaly Detection and Segmentation MVTEC 3D-AD Back to the Feature: Classical 3D Features are (Almost) All You Need for 3D Anomaly Detection (HoG) Segmentation AUROC 0.930 #7 of 13 Archive leaderboard report
Depth Anomaly Detection and Segmentation MVTEC 3D-AD Back to the Feature: Classical 3D Features are (Almost) All You Need for 3D Anomaly Detection (Depth iNet) Detection AUROC 0.675 #8 of 13 Archive leaderboard report
Depth Anomaly Detection and Segmentation MVTEC 3D-AD Back to the Feature: Classical 3D Features are (Almost) All You Need for 3D Anomaly Detection (Depth iNet) Segmentation AUPRO 0.755 #8 of 13 Archive leaderboard report
Depth Anomaly Detection and Segmentation MVTEC 3D-AD Back to the Feature: Classical 3D Features are (Almost) All You Need for 3D Anomaly Detection (Depth iNet) Segmentation AUROC 0.930 #8 of 13 Archive leaderboard report
Depth Anomaly Detection and Segmentation MVTEC 3D-AD Back to the Feature: Classical 3D Features are (Almost) All You Need for 3D Anomaly Detection (NSA) Detection AUROC 0.696 #9 of 13 Archive leaderboard report
Depth Anomaly Detection and Segmentation MVTEC 3D-AD Back to the Feature: Classical 3D Features are (Almost) All You Need for 3D Anomaly Detection (NSA) Segmentation AUPRO 0.5572 #9 of 13 Archive leaderboard report
Depth Anomaly Detection and Segmentation MVTEC 3D-AD Back to the Feature: Classical 3D Features are (Almost) All You Need for 3D Anomaly Detection (NSA) Segmentation AUROC 0.817 #9 of 13 Archive leaderboard report
Depth Anomaly Detection and Segmentation MVTEC 3D-AD Back to the Feature: Classical 3D Features are (Almost) All You Need for 3D Anomaly Detection (RaW) Detection AUROC 0.573 #10 of 13 Archive leaderboard report
Depth Anomaly Detection and Segmentation MVTEC 3D-AD Back to the Feature: Classical 3D Features are (Almost) All You Need for 3D Anomaly Detection (RaW) Segmentation AUPRO 0.442 #10 of 13 Archive leaderboard report
Depth Anomaly Detection and Segmentation MVTEC 3D-AD Back to the Feature: Classical 3D Features are (Almost) All You Need for 3D Anomaly Detection (RaW) Segmentation AUROC 0.771 #10 of 13 Archive leaderboard report
RGB+3D Anomaly Detection and Segmentation MVTEC 3D-AD Back to the Feature: Classical 3D Features are (Almost) All You Need for 3D Anomaly Detection (BTF) Detection AUCROC 0.865 #6 of 9 Archive leaderboard report
RGB+3D Anomaly Detection and Segmentation MVTEC 3D-AD Back to the Feature: Classical 3D Features are (Almost) All You Need for 3D Anomaly Detection (BTF) Segmentation AUCROC 0.992 #6 of 9 Archive leaderboard report
RGB+3D Anomaly Detection and Segmentation MVTEC 3D-AD Back to the Feature: Classical 3D Features are (Almost) All You Need for 3D Anomaly Detection (BTF) Segmentation AUPRO 0.959 #6 of 9 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

Introduced by this paper: BTF

BTF

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