Browse › Computer Vision › Action Recognition › UCF101

Action Recognition archive 2025-07-28

UCF101 Benchmark (Action Recognition)

91 rows 58 with code listed 3 metrics Dataset page

Syntology Since the archive (Syntology): none placed; no newer paper judged to report on this board so far.

Action Recognition is a computer vision task that involves recognizing human actions in videos or images. The goal is to classify and categorize the actions being performed in the video or image into a predefined set of action classes.

In the video domain, it is an open question whether training an action classification network on a sufficiently large dataset, will give a similar boost in performance when applied to a different temporal task or dataset. The challenges of building video datasets has meant that most popular benchmarks for action recognition are small, having on the order of 10k videos.

Please note some benchmarks may be located in the Action Classification or Video Classification tasks, e.g. Kinetics-400.

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

Over time archive 2025-07-28

Direction inferred from the metric name, not from the archive: 3-fold Accuracy (higher is better), Accuracy (higher is better), Accuracy 20%Test (higher is better). Points are placed at the row's paper date, not an evaluation date; 90 of 91 rows carry one.

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

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 FTP-UniFormerV2-L/14 99.7 – Paper – 2024 no code linked report
2 VideoMAE V2-g 99.6 ✓ Paper Code 2023 official (archive's flag): 2 ran · 2 ran (of which 2 constructed an object rather than computing a result; 2 with no instrument failure: 0 honoured, 0 violated, 2 with no contract checked; 0 where Syntology's instrument failed) · 4 unverified; every one of the 2 samples that ran constructed an object rather than computing a result report
3 OmniVec 99.6 ✓ Paper – 2023 no code linked report
4 OmniVec2 99.6 ✓ Paper – 2024 no code linked report
5 BIKE 98.8 ✓ Paper Code 2022 linked, not harvested report
6 SMART 98.64 – Paper – 2020 no code linked report
7 OmniSource (SlowOnly-8x8-R101-RGB + I3D-Flow) 98.6 ✓ Paper Code 2020 official (archive's flag): 10 ran · 10 ran (of which 0 constructed an object rather than computing a result; 10 with no instrument failure: 0 honoured, 0 violated, 10 with no contract checked; 0 where Syntology's instrument failed) · 3 unverified report
8 PERF-Net (multi-distilled S3D) 98.6 – Paper – 2020 no code linked report
9 ZeroI2V ViT-L/14 98.6 ✓ Paper Code 2023 official (archive's flag): 1 ran · 1 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 1 violated, 0 with no contract checked; 0 where Syntology's instrument failed) · 2 unverified report
10 LGD-3D Two-stream 98.2 – Paper – 2019 no code linked report
11 Text4Vis 98.2 – Paper Code 2022 linked, not harvested report
12 Two-Stream I3D (Imagenet+Kinetics pre-training) 98.0 – Paper Code 2017 16 ran (of which 7 constructed an object rather than computing a result; 10 with no instrument failure: 0 honoured, 0 violated, 10 with no contract checked; 6 where Syntology's instrument failed) · 12 unverified report
13 MARS+RGB+Flow (64 frames, Kinetics pretrained) 97.8 ✓ Paper Code 2019 linked, not harvested report
14 HATNet (32 frames) 97.8 – Paper Code 2019 linked, not harvested report
15 Two-Stream I3D (Kinetics pre-training) 97.8 – Paper Code 2017 16 ran (of which 7 constructed an object rather than computing a result; 10 with no instrument failure: 0 honoured, 0 violated, 10 with no contract checked; 6 where Syntology's instrument failed) · 12 unverified report
16 BubbleNET 97.62 ✓ Paper – 2020 no code linked report
17 D3D + D3D 97.6 – Paper Code 2018 linked, not harvested report
18 BQN 97.6 – Paper Code 2021 linked, not harvested report
19 CCS + TSN (ImageNet+Kinetics pretrained) 97.4 ✓ Paper – 2019 no code linked report
20 R[2+1]D-TwoStream (Kinetics pretrained) 97.3 ✓ Paper Code 2017 community repositories only · 1 ran (of which 0 constructed an object rather than computing a result; 0 with no instrument failure: 0 honoured, 0 violated, 0 with no contract checked; 1 where Syntology's instrument failed) · 3 unverified report
21 Multi-stream I3D 97.2 – Paper – 2019 no code linked report
22 CA2ST(B/16) 97.2 – Paper – 2025 no code linked report
23 Hidden Two-Stream 97.1 – Paper Code 2017 linked, not harvested report
24 D3D (Kinetics-600 pretraining) 97.1 ✓ Paper Code 2018 linked, not harvested report
25 AMD(ViT-B/16) 97.1 ✓ Paper – 2023 no code linked report
26 D3D (Kinetics-400 pretraining) 97 ✓ Paper Code 2018 linked, not harvested report
27 LGD-3D RGB 97 – Paper – 2019 no code linked report
28 STAM-32 (ImageNet/Kinetics pretraining) 97 ✓ Paper Code 2021 official: harvested, nothing ran · 0 ran · 1 unverified report
29 FASTER32 96.9 – Paper – 2019 no code linked report
30 R[2+1]D-RGB (Kinetics pretrained) 96.8 ✓ Paper Code 2017 community repositories only · 1 ran (of which 0 constructed an object rather than computing a result; 0 with no instrument failure: 0 honoured, 0 violated, 0 with no contract checked; 1 where Syntology's instrument failed) · 3 unverified report
31 S3D-G (ImageNet, Kinetics-400 pretrained) 96.8 ✓ Paper Code 2017 linked, not harvested report
32 LGD-3D Flow 96.8 – Paper – 2019 no code linked report
33 Flow-I3D (Imagenet+Kinetics pre-training) 96.7 ✓ Paper Code 2017 16 ran (of which 7 constructed an object rather than computing a result; 10 with no instrument failure: 0 honoured, 0 violated, 10 with no contract checked; 6 where Syntology's instrument failed) · 12 unverified report
34 VidTr-L 96.7 – Paper – 2021 no code linked report
35 CMA iter1-S 96.5 – Paper – 2019 no code linked report
36 Flow-I3D (Kinetics pre-training) 96.5 ✓ Paper Code 2017 16 ran (of which 7 constructed an object rather than computing a result; 10 with no instrument failure: 0 honoured, 0 violated, 10 with no contract checked; 6 where Syntology's instrument failed) · 12 unverified report
37 I3D RGB + DMC-Net (I3D) 96.5 – Paper – 2019 no code linked report
38 A2-Net (ResNet-50) 96.4 – Paper – 2018 no code linked report
39 MF-Net, RGB only (ImageNet+Kinetics pretrained) 96.0 ✓ Paper – 2018 no code linked report
40 Optical Flow Guided Feature 96 – Paper Code 2017 linked, not harvested report
41 MARS+RGB+Flow (16 frames) 95.8 – Paper Code 2019 linked, not harvested report
42 Prob-Distill 95.7 – Paper – 2019 no code linked report
43 RGB-I3D (Imagenet+Kinetics pre-training) 95.6 ✓ Paper Code 2017 16 ran (of which 7 constructed an object rather than computing a result; 10 with no instrument failure: 0 honoured, 0 violated, 10 with no contract checked; 6 where Syntology's instrument failed) · 12 unverified report
44 R[2+1]D-Flow (Kinetics pretrained) 95.5 ✓ Paper Code 2017 community repositories only · 1 ran (of which 0 constructed an object rather than computing a result; 0 with no instrument failure: 0 honoured, 0 violated, 0 with no contract checked; 1 where Syntology's instrument failed) · 3 unverified report
45 TVNet+IDT 95.4 – Paper Code 2018 0 ran · 2 unverified report
46 TesNet (ImageNet pretrained) 95.2 ✓ Paper – 2020 no code linked report
47 I3D-LSTM 95.1 – Paper Code 2019 linked, not harvested report
48 RGB-I3D (Kinetics pre-training) 95.1 ✓ Paper Code 2017 16 ran (of which 7 constructed an object rather than computing a result; 10 with no instrument failure: 0 honoured, 0 violated, 10 with no contract checked; 6 where Syntology's instrument failed) · 12 unverified report
49 R[2+1]D-TwoStream (Sports-1M pretrained) 95 ✓ Paper Code 2017 community repositories only · 1 ran (of which 0 constructed an object rather than computing a result; 0 with no instrument failure: 0 honoured, 0 violated, 0 with no contract checked; 1 where Syntology's instrument failed) · 3 unverified report
50 X3D MobileNet-V3 LGD-GC 94.85 ✓ Paper Code 2021 linked, not harvested report
51 ST-ResNet + IDT 94.6 – Paper Code 2016 linked, not harvested report
52 ResNeXt-101 (64f) 94.5 – Paper Code 2017 official: no sample here; runs from other or unrecorded repositories · 7 ran (of which 0 constructed an object rather than computing a result; 0 with no instrument failure: 0 honoured, 0 violated, 0 with no contract checked; 7 where Syntology's instrument failed) · 1 unverified report
53 R-STAN-101 94.5 – Paper – 2019 no code linked report
54 TSN+TSM 94.3 – Paper – 2018 no code linked report
55 ARTNet w/ TSN 94.3 – Paper Code 2017 linked, not harvested report
56 Temporal Segment Networks 94.2 – Paper Code 2016 community repositories only · 10 ran (of which 0 constructed an object rather than computing a result; 10 with no instrument failure: 2 honoured, 0 violated, 8 with no contract checked; 0 where Syntology's instrument failed) · 14 unverified report
57 TS-LSTM 94.1 – Paper Code 2017 linked, not harvested report
58 SVT 93.7 – Paper Code 2021 official: harvested, nothing ran · 0 ran · 1 unverified report
59 R[2+1]D-RGB (Sports-1M pretrained) 93.6 ✓ Paper Code 2017 community repositories only · 1 ran (of which 0 constructed an object rather than computing a result; 0 with no instrument failure: 0 honoured, 0 violated, 0 with no contract checked; 1 where Syntology's instrument failed) · 3 unverified report
60 Two-stream I3D 93.4 – Paper Code 2017 16 ran (of which 7 constructed an object rather than computing a result; 10 with no instrument failure: 0 honoured, 0 violated, 10 with no contract checked; 6 where Syntology's instrument failed) · 12 unverified report
61 R[2+1]D-Flow (Sports-1M pretrained) 93.3 ✓ Paper Code 2017 community repositories only · 1 ran (of which 0 constructed an object rather than computing a result; 0 with no instrument failure: 0 honoured, 0 violated, 0 with no contract checked; 1 where Syntology's instrument failed) · 3 unverified report
62 VIMPAC 92.7 – Paper Code 2021 official (archive's flag): 1 ran · 1 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 1 violated, 0 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified report
63 S:VGG-16, T:VGG-16 (ImageNet pretrain) 92.5 ✓ Paper Code 2016 linked, not harvested report
64 DMC-Net (I3D) 92.3 – Paper – 2019 no code linked report
65 two-in-one two stream 92 – Paper Code 2019 linked, not harvested report
66 LTC 91.7 – Paper Code 2016 linked, not harvested report
67 R-STAN-50 91.5 – Paper – 2019 no code linked report
68 TDD + IDT 91.5 – Paper Code 2015 linked, not harvested report
69 Very deep two-stream ConvNet 91.4 – Paper Code 2015 linked, not harvested report
70 3D ResNeXt-101 + Confidence Distillation 91.2 – Paper – 2021 no code linked report
71 MR Two-Sream R-CNN 91.1 – Paper – 2016 no code linked report
72 Dynamic Image Networks + IDT 89.1 – Paper Code 2016 linked, not harvested report
73 Two-stream+LSTM 88.6 – Paper Code 2015 3 ran (of which 0 constructed an object rather than computing a result; 3 with no instrument failure: 0 honoured, 0 violated, 3 with no contract checked; 0 where Syntology's instrument failed) · 1 unverified report
74 P3D (ImageNet + Sports1M) 88.6 ✓ Paper Code 2017 linked, not harvested report
75 Two-Stream (ImageNet pretrained) 88.0 ✓ Paper Code 2014 community repositories only · 5 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; 1 where Syntology's instrument failed) · 2 unverified report
76 MV-CNN 86.4 – Paper Code 2016 linked, not harvested report
77 Dynamics 2 for DenseNet-201 Transformer 86.1 – Paper Code 2023 linked, not harvested report
78 R(2+1)D-18 (DistInit pretraining) 85.8 – Paper – 2019 no code linked report
79 Res3D 85.8 – Paper Code 2017 linked, not harvested report
80 ActionFlowNet 83.9 – Paper – 2016 no code linked report
81 C3D 82.3 – Paper Code 2014 linked, not harvested report
82 HalluciNet (ResNet-50) 79.83 – Paper Code 2019 linked, not harvested report
83 R[2+1]D (VideoMoCo) 78.7 – Paper Code 2021 official (archive's flag): 1 ran · 1 ran (of which 1 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified; the one sample that ran constructed an object rather than computing a result report
84 3D-ResNet-18 (VideoMoCo) 74.1 – Paper Code 2021 official (archive's flag): 1 ran · 1 ran (of which 1 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified; the one sample that ran constructed an object rather than computing a result report
85 Slow Fusion + Finetune top 3 layers 65.4 ✓ Paper Code 2014 linked, not harvested report
86 MLGCN 63.27 – Paper – 2019 no code linked report
87 CD-UAR 42.5 – Paper – 2018 no code linked report
88 SL 35.2 – – – not matched report
89 I3D + PoTion 29.3 – Paper – 2018 no code linked report
90 R3D-18 73.16 – Paper Code 2022 linked, not harvested report
91 ResNet50 98.05 – Paper Code 2022 linked, not harvested report

All 91 rows shown. 90 link to a paper page on this site; 31 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 ran (of which C constructed an object rather than computing a result; K with no instrument failure: H honoured, V violated, P with no contract checked; I where Syntology's instrument failed) · U unverified”: of the code samples Syntology harvested from repositories linked to that row's paper (joined by arXiv id), N executed on a synthesized input and U are unverified (harvested, no recorded run). Of the N, C constructed an object rather than computing a result (a part of N, never taken away from it), K ran with no instrument failure (honoured or violated the contract Syntology drafted, or ran with no contract checked) and I are failures of Syntology's instrument, not of the code (a wrong contract draft, or a contract check that could not 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). 26 rows have a graph line, from 14 distinct papers. Rows whose paper has a sample that ran: 23 (16 with a run with no instrument failure, 7 where every run was a failure of Syntology's instrument). Papers: 11 with a sample that ran: 9 with a run with no instrument failure, 2 where every run was a failure of Syntology's instrument. Whose code ran. When the archive marks a repository official for the row's paper, the cell starts with that repository's state (“official” is the archive's flag, not a verdict on who wrote the code): “official (archive's flag): N ran” counts the samples from it that ran, instrument failures included; “official: harvested, nothing ran”; when it has no sample for this paper and other samples ran, “community repositories only” if every one came from a repository the archive lists for the paper but neither marks official nor records as mentioned in it, or one Syntology harvested outside the archive's links, and “official: no sample here; runs from other or unrecorded repositories” if any came from a repository named in the paper or found in its text by Syntology, or has no repository recorded; “official: harvested for another paper” (no sample from it is linked to this paper) or “official: not harvested”. No state is printed when the archive marks none official. The cell's tooltip names the repositories the samples that ran came from. Of the 26 rows with a graph line, 17 have an archive-official repository: on 6 a sample from it ran (6 with a run with no instrument failure), on 8 the samples that ran came from community repositories only, and on 1 from other or unrecorded repositories. The filters above hide rows and never renumber them; the first two count, by default, only a run with no instrument failure, and offer any run, instrument failures included. Counting each paper once, of 103 samples: 57 ran (of which 10 constructed an object rather than computing a result; 42 with no instrument failure: 2 honoured, 2 violated, 38 with no contract checked; 15 where Syntology's instrument failed) · 46 unverified. Separately, 21 of those 103 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. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record. Per-sample status is on the paper page; each Ran cell links to that paper's sample list. For agents: get_harvested_code_for_paper(arxiv_id) lists each paper's samples; how to connect.

Since the archive Syntology

Of the 8,888 newer papers whose tables Syntology has read (8,886 checked in full and 2 with a leaderboard still awaiting a verdict), 11 had this board among the leaderboards the extractor considered for their tables (it keeps at most 10 boards per paper and 4 per dataset; the lists of 1,585 of the 8,888 were already full without this board, so if it was crowded out of them it is not counted), and none proposed a result, so none await a verdict and none were placed. Syntology has checked 8,886 of the 18,637 papers on this site that are newer than the archive (for 3,717 of them no archive leaderboard matched the paper's tables, so there was nothing further to check); 733 were read and have nothing to place (arXiv has no HTML version of the paper, or that version has no tables), 41 could not be read (the extractor's reply could not be parsed), and results from the other 8,977 appear after they are checked.

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