Browse › Computer Vision › Image Classification › STL-10

Image Classification archive 2025-07-28

STL-10 Benchmark (Image Classification)

117 rows 66 with code listed 3 metrics Dataset page

Image Classification is a fundamental task in vision recognition that aims to understand and categorize an image as a whole under a specific label. Unlike object detection, which involves classification and location of multiple objects within an image, image classification typically pertains to single-object images. When the classification becomes highly detailed or reaches instance-level, it is often referred to as image retrieval, which also involves finding similar images in a large database.

Source: Metamorphic Testing for Object Detection Systems

The archive carries no text for this table; the description above is the archive's text for the task Image Classification. archive 2025-07-28

Over time archive 2025-07-28

The chart needs JavaScript; the table below carries every value.

Direction inferred from the metric name, not from the archive: FLOPS (lower is better), PARAMS (lower is better). Not inferred (points only, no best-so-far line): Percentage correct. Points are placed at the row's paper date; 117 of 117 rows carry one.

Results archive 2025-07-28

Archive rows end at the archive snapshot, 2025-07-28: no result published after that date is in this table. Rank is the archive's row order at that snapshot; not re-ranked here. Metric values are the archive's strings. Column headers sort the table in your browser; each row keeps its archive rank.

Paper Code Ran Syntology Report
1 µ2Net+ (ViT-L/16) 99.64 ✓ Paper Code 2022 linked, not harvested report
2 kNN-CLIP 99.6 ✓ Paper – 2022 no code linked report
3 Wide-ResNet-101 (Spinal FC) 98.66 ✓ Paper Code 2020 linked, not harvested report
4 CN(d=128) 98.45 ✓ Paper – 2021 no code linked report
5 CN(d=64) 98.36 ✓ Paper – 2021 no code linked report
6 NSRL+CN(d=128) 98.36 ✓ Paper – 2021 no code linked report
7 NSRL+CN(d=32) 98.34 ✓ Paper – 2021 no code linked report
8 NSRL+CN(d=64) 98.24 ✓ Paper – 2021 no code linked report
9 CN(d=32) 98.17 ✓ Paper – 2021 no code linked report
10 NAT-M4 97.9573M7.5M ✓ Paper Code 2020 0 of 1 ran · 1 unverified report
11 NAT-M3 97.8436M7.5M ✓ Paper Code 2020 0 of 1 ran · 1 unverified report
12 SEER (RegNet10B) 97.310000M ✓ Paper Code 2022 linked, not harvested report
13 NAT-M2 97.2303M5.1M ✓ Paper Code 2020 0 of 1 ran · 1 unverified report
14 iGPT-L 97.1 ✓ Paper Code 2020 linked, not harvested report
15 NAT-M1 96.7240M4.4M ✓ Paper Code 2020 0 of 1 ran · 1 unverified report
16 EnAET 95.48 ✓ Paper Code 2019 linked, not harvested report
17 VGG-19bn 95.44 ✓ Paper Code 2020 linked, not harvested report
18 Diffusion Classifier (zero-shot) 95.4 – Paper Code 2023 2 of 2 ran · 0 unverified report
19 FixMatch (CTA) 94.83 – Paper Code 2020 50 of 74 ran · 24 unverified report
20 ReMixMatch 94.77 – Paper Code 2020 50 of 74 ran · 24 unverified report
21 AMDIM 94.5 – Paper Code 2019 8 of 11 ran · 3 unverified report
22 MixMatch 94.41 – Paper Code 2019 39 of 63 ran · 24 unverified report
23 AMDIM-L 94.2 ✓ Paper Code 2020 linked, not harvested report
24 ReMixMatch (K=4) 93.82 – Paper Code 2019 1 of 4 ran · 3 unverified report
25 AMDIM 93.80 – Paper Code 2020 0 of 2 ran · 2 unverified report
26 ReMixMatch (K=1) 93.23 – Paper Code 2019 1 of 4 ran · 3 unverified report
27 MP* 93.19 – Paper – 2020 no code linked report
28 UDA 92.34 – Paper Code 2020 50 of 74 ran · 24 unverified report
29 YADIM 92.15 – Paper Code 2020 0 of 2 ran · 2 unverified report
30 FixMatch (RA) 92.02 – Paper Code 2020 50 of 74 ran · 24 unverified report
31 NSGANetV2 92.0 – Paper Code 2020 2 of 8 ran · 6 unverified report
32 SESN 91.49 – Paper Code 2019 0 of 8 ran · 8 unverified report
33 Harmonic WRN-16-8 90.45 – Paper Code 2019 linked, not harvested report
34 wrn16/8 D8 D4 D1 90.20 – Paper Code 2019 linked, not harvested report
35 DLME (ResNet-50, linear) 90.1 – Paper Code 2022 1 of 1 ran · 0 unverified report
36 MixMatch 89.82 – Paper Code 2019 1 of 4 ran · 3 unverified report
37 MixMatch 89.59 – Paper Code 2020 50 of 74 ran · 24 unverified report
38 wrn16/8* D8 D4 D1 89.43 – Paper Code 2019 linked, not harvested report
39 wrn16/8* D1 D1 D1 88.95 – Paper Code 2019 linked, not harvested report
40 wrn16/8 D1 D1 D1 88.83 – Paper Code 2019 linked, not harvested report
41 IIC 88.8 – Paper Code 2018 2 of 18 ran · 16 unverified report
42 IIC 88.80 – Paper Code 2019 39 of 63 ran · 24 unverified report
43 SOPCNN 88.08 – Paper Code 2020 linked, not harvested report
44 TS 88.03 – Paper – 2020 no code linked report
45 CutOut 87.36 – Paper Code 2019 39 of 63 ran · 24 unverified report
46 Cutout 87.26 – Paper Code 2017 21 of 24 ran · 3 unverified report
47 wrn16/8 87.26 – Paper Code 2019 linked, not harvested report
48 Hamiltonian 85.5 – Paper Code 2017 linked, not harvested report
49 ResNet-18+MM+FRL 85.42 – Paper – 2020 no code linked report
50 MidPoint 84.6 – Paper Code 2017 linked, not harvested report
51 cosine function 84.38 – Paper – 2018 no code linked report
52 HybridNet 84.10 – Paper – 2018 no code linked report
53 Leapfrog 83.7 – Paper Code 2017 linked, not harvested report
54 skewing 83.47 – Paper – 2018 no code linked report
55 elastic distortion(2) 83.45 – Paper – 2018 no code linked report
56 PSLR-knn 83.2 – Paper – 2020 no code linked report
57 elastic distortion(1) 83.00 – Paper – 2018 no code linked report
58 ResNet baseline 82.00 – Paper – 2018 no code linked report
59 Greedy InfoMax (GIM) 81.9 – Paper Code 2019 1 of 4 ran · 3 unverified report
60 rotation 81.45 – Paper – 2018 no code linked report
61 ResNet18(BN, 4) 81.04 – Paper – 2020 no code linked report
62 VGG8B + LocalLearning + CO 80.75 – Paper Code 2019 linked, not harvested report
63 ResNet18(GN, 4) 79.3 – Paper – 2020 no code linked report
64 PSLR-Linear 78.8 – Paper – 2020 no code linked report
65 ResNet18(BN, 128) 78.65 – Paper – 2020 no code linked report
66 Mean Teacher 78.57 – Paper Code 2020 50 of 74 ran · 24 unverified report
67 CPC† 78.36 – Paper Code 2020 0 of 2 ran · 2 unverified report
68 Hamiltonian 78.3 – Paper Code 2018 linked, not harvested report
69 CC-GAN² 77.8 – Paper Code 2016 0 of 1 ran · 1 unverified report
70 CC-GAN 77.80 – Paper Code 2019 1 of 4 ran · 3 unverified report
71 Parabolic 77.0 – Paper Code 2018 linked, not harvested report
72 Scat + WRN 20-8 76.6 – Paper Code 2017 2 of 2 ran · 0 unverified report
73 ResNet18(EBN, 4) 76.49 – Paper – 2020 no code linked report
74 Exemplar CNN 75.7 – Paper Code 2017 2 of 2 ran · 0 unverified report
75 ResNet18(EBN, 128) 75.57 – Paper – 2020 no code linked report
76 Stacked what-where AE 74.33 – Paper Code 2017 2 of 2 ran · 0 unverified report
77 SWWAE 74.33 – Paper – 2018 no code linked report
78 SWWAE 74.30 – Paper Code 2019 1 of 4 ran · 3 unverified report
79 SWWAE 74.3 – Paper Code 2015 linked, not harvested report
80 Second-order 74.3 – Paper Code 2018 linked, not harvested report
81 Convolutional Clustering 74.1 – Paper – 2015 no code linked report
82 Π-Model 73.77 – Paper Code 2020 50 of 74 ran · 24 unverified report
83 Discriminative Unsupervised Feature Learning with Convolutional Neural Networks 72.8 – Paper Code 2014 linked, not harvested report
84 ResNet18(GN, 128) 72.66 – Paper – 2020 no code linked report
85 Pseudo-Labeling 72.01 – Paper Code 2020 50 of 74 ran · 24 unverified report
86 Entropy 71.65 – Paper – 2020 no code linked report
87 BDW 71.12 – Paper – 2020 no code linked report
88 MP 71.05 – Paper – 2020 no code linked report
89 WaveMixLite-256/7 70.88 – Paper Code 2022 linked, not harvested report
90 CNN 70.7 – Paper Code 2017 2 of 2 ran · 0 unverified report
91 An Analysis of Unsupervised Pre-training in Light of Recent Advances 70.2 – Paper Code 2014 linked, not harvested report
92 Multi-Task Bayesian Optimization 70.1 – Paper Code 2013 linked, not harvested report
93 NN-Weighter 69.15 – Paper – 2020 no code linked report
94 Accuracy Monitoring 68.62 – Paper – 2020 no code linked report
95 C-SVDDNet 68.2 – Paper – 2014 no code linked report
96 RotNet 68.19 – Paper – 2020 no code linked report
97 DFF Committees 68 – Paper – 2014 no code linked report
98 Hierarchical Matching Pursuit (HMP) 64.6 – Paper Code 2017 2 of 2 ran · 0 unverified report
99 L2RW 63.13 – Paper – 2020 no code linked report
100 Discriminative Learning of Sum-Product Networks 62.3 – Paper – 2011 no code linked report
101 CKN 62.3 – Paper – 2014 no code linked report
102 S-CNN 61.94 – Paper – 2016 no code linked report
103 Simulated Fixations 61 – Paper Code 2020 0 of 2 ran · 2 unverified report
104 No more meta-parameter tuning in unsupervised sparse feature learning 61 – Paper – 2014 no code linked report
105 Convolutional K-means Network 60.2 – Paper Code 2017 2 of 2 ran · 0 unverified report
106 Receptive Fields 60.1 – Paper – 2007 no code linked report
107 PWD 59.45 – Paper – 2020 no code linked report
108 GVD 59.33 – Paper – 2020 no code linked report
109 VR 59.13 – Paper – 2020 no code linked report
110 Core SET 58.93 – Paper – 2020 no code linked report
111 GE 58.84 – Paper – 2020 no code linked report
112 DFAL 58.81 – Paper – 2020 no code linked report
113 Random 58.15 – Paper – 2020 no code linked report
114 BALD-MCD 57.35 – Paper – 2020 no code linked report
115 Sign-symmetry 57.32 – Paper Code 2015 linked, not harvested report
116 M2-PWD 57.31 – Paper – 2020 no code linked report
117 soft ica 52.9 – Paper – 2011 no code linked report

All 117 rows shown. 117 link to a paper page on this site; 18 are marked as using additional training data in the archive. No GitHub stars are tracked; "Code" is the first repository the archive lists for the row. The archive carries no row tags, review links or community-submitted rows for this table; none are shown. archive 2025-07-28

Syntology Ran reads "N of M ran · U unverified": of the M code samples Syntology harvested from repositories linked to that row's paper (joined by arXiv id), N executed on a synthesized input and the other U = M−N are unverified (harvested, no recorded run). It counts code from repositories linked to that row's paper, not this result: the row's number was not reproduced and nothing here is a correctness claim. The other cell texts mean no graph line for the row: "linked, not harvested" (the archive links code, Syntology has not harvested it), "no code linked" (no code link in the archive), "not matched" (the row's paper URL matched no paper on this site). 39 rows have a graph line, from 15 distinct papers; 29 rows (11 papers) have at least one sample that ran. Counting each paper once: Syntology ran 129 of 223 samples; 94 unverified. Separately, 60 of those 223 are pointer-only (licence): the site points at that code rather than redistributing it, a licence property recorded for ran and unverified samples alike; each cell's tooltip carries the row's own pointer-only count. Read from the graph 2026-09-24. Per-sample status is on the paper page.

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