</> Code with Papers by Syntology

Browse › Computer Vision › Image Classification › iNaturalist 2018

Image Classification archive 2025-07-28

iNaturalist 2018 Benchmark (Image Classification)

60 rows 53 with code listed 2 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: Top-1 Accuracy (higher is better), Number of params (lower is better). Points are placed at the row's paper date; 60 of 60 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 OmniVec2 94.6 ✓ Paper – 2024 no code linked report
2 OmniVec 93.8 ✓ Paper – 2023 no code linked report
3 InternImage-H 92.6% ✓ Paper Code 2022 2 of 4 ran · 2 unverified report
4 MAWS (ViT-2B) 91.3% ✓ Paper Code 2023 linked, not harvested report
5 MetaFormer (MetaFormer-2,384,extra_info) 88.7% ✓ Paper Code 2022 1 of 14 ran · 13 unverified report
6 Hiera-H (448px) 87.3% ✓ Paper Code 2023 0 of 6 ran · 6 unverified report
7 MAE (ViT-H, 448) 86.8% ✓ Paper Code 2021 71 of 137 ran · 66 unverified report
8 SWAG (ViT H/14) 86.0% ✓ Paper Code 2022 linked, not harvested report
9 SEER (RegNet10B - finetuned - 384px) 84.7% ✓ Paper Code 2022 linked, not harvested report
10 MetaFormer (MetaFormer-2,384) 84.3% ✓ Paper Code 2022 1 of 14 ran · 13 unverified report
11 OMNIVORE (Swin-L) 84.1% ✓ Paper Code 2022 2 of 2 ran · 0 unverified report
12 RDNet-L (224 res, IN-1K pretrained) 81.8%186M – Paper Code 2024 3 of 3 ran · 0 unverified report
13 RegNet-8GF 81.2% ✓ Paper – 2020 no code linked report
14 VL-LTR (ViT-B-16) 81.0% ✓ Paper Code 2021 linked, not harvested report
15 µ2Net+ (ViT-L/16) 80.97 – Paper Code 2022 linked, not harvested report
16 RDNet-B (224 res, IN-1K pretrained) 80.587M – Paper Code 2024 3 of 3 ran · 0 unverified report
17 MixMIM-L 80.3% – Paper Code 2022 linked, not harvested report
18 DeiT-B 79.5% ✓ Paper Code 2020 12 of 19 ran · 7 unverified report
19 CeiT-S (384 finetune resolution) 79.4% ✓ Paper Code 2021 9 of 11 ran · 2 unverified report
20 RDNet-S (224 res, IN-1K pretrained) 79.150M – Paper Code 2024 3 of 3 ran · 0 unverified report
21 GPaCo (ResNet-152) 78.1% – Paper Code 2022 linked, not harvested report
22 CaiT-M-36 U 224 78% ✓ Paper Code 2021 5 of 11 ran · 6 unverified report
23 MixMIM-B 77.5% – Paper Code 2022 linked, not harvested report
24 RDNet-T (224 res, IN-1K pretrained) 77.024M – Paper Code 2024 3 of 3 ran · 0 unverified report
25 GPaCo (ResNet-50) 75.4% – Paper Code 2022 linked, not harvested report
26 CBD-ENS (ResNet-101) 75.3% – Paper Code 2021 linked, not harvested report
27 ViT-L (attn finetune) 75.3% ✓ Paper Code 2022 1 of 1 ran · 0 unverified report
28 PaCo(ResNet-152) 75.2% – Paper Code 2021 1 of 3 ran · 2 unverified report
29 VL-LTR (ResNet-50) 74.6% ✓ Paper Code 2021 linked, not harvested report
30 BS-CMO (ResNet-50) 74.0% – Paper Code 2021 3 of 5 ran · 2 unverified report
31 CBD-ENS (ResNet-50) 73.6% – Paper Code 2021 linked, not harvested report
32 CeiT-S 73.3% – Paper Code 2021 9 of 11 ran · 2 unverified report
33 TADE (ResNet-50) 72.9% – Paper Code 2021 4 of 6 ran · 2 unverified report
34 CeiT-T (384 finetune resolution) 72.2% – Paper Code 2021 9 of 11 ran · 2 unverified report
35 RIDE (ResNet-50) 72.2% – Paper Code 2020 1 of 1 ran · 0 unverified report
36 ResNeXt-101 (SAMix) 70.54% – Paper Code 2021 linked, not harvested report
37 ResNeXt-101 (AutoMix) 70.49% – Paper Code 2021 4 of 6 ran · 2 unverified report
38 LADE 70.0% – Paper Code 2020 linked, not harvested report
39 ResNet-50 69.8% – Paper – 2020 no code linked report
40 ResNet-152 69.08% – Paper – 2020 no code linked report
41 ResNet-152 69.05% – Paper Code 2019 7 of 27 ran · 20 unverified report
42 MetaSAug 68.75% – Paper Code 2021 2 of 3 ran · 1 unverified report
43 ResNet-101 68.39% – Paper – 2020 no code linked report
44 ResNet-101 67.98% – Paper Code 2019 7 of 27 ran · 20 unverified report
45 LeViT-384 66.9% – Paper Code 2021 23 of 30 ran · 7 unverified report
46 LeViT-256 66.2% – Paper Code 2021 23 of 30 ran · 7 unverified report
47 ResNet-50 65.91% – Paper – 2020 no code linked report
48 ResNet-50 (SAMix) 64.84% – Paper Code 2021 linked, not harvested report
49 ResNet-50 (AutoMix) 64.73% – Paper Code 2021 4 of 6 ran · 2 unverified report
50 CeiT-T 64.3% – Paper Code 2021 9 of 11 ran · 2 unverified report
51 ResMLP-24 64.3 – Paper Code 2021 2 of 7 ran · 5 unverified report
52 ResNet-50 64.16% – Paper Code 2019 7 of 27 ran · 20 unverified report
53 LeViT-192 60.4% – Paper Code 2021 23 of 30 ran · 7 unverified report
54 Inception-V3 60.20% – Paper Code 2017 linked, not harvested report
55 ResMLP-12 60.2 – Paper Code 2021 2 of 7 ran · 5 unverified report
56 LeViT-128S 55.2% – Paper Code 2021 23 of 30 ran · 7 unverified report
57 LeViT-128 54% – Paper Code 2021 23 of 30 ran · 7 unverified report
58 ResNet-50 49.7% – Paper Code 2019 linked, not harvested report
59 ResNet-50 48.6 – Paper Code 2020 13 of 17 ran · 4 unverified report
60 Barlow Twins (ResNet-50) 46.5 – Paper Code 2021 21 of 26 ran · 5 unverified report

All 60 rows shown. 60 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). 36 rows have a graph line, from 21 distinct papers; 35 rows (20 papers) have at least one sample that ran. Counting each paper once: Syntology ran 187 of 339 samples; 152 unverified. Separately, 98 of those 339 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