# Action Recognition on SSV2 & K400

Source: [Action Recognition on SSV2 & K400](<https://www.sota2.com/research/sota/action-recognition-on-ssv2-and-k400>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-02-26T05:46:23.779Z

Evaluation of Action Recognition performance across the combined SSV2 and Kinetics-400 datasets, utilizing metrics such as the Harmonic Mean.

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

| Scope | Recorded value |
| --- | --- |
| Task | Action Recognition |
| Dataset (including any stated split) | SSV2 & K400 |
| Metric | Harmonic Mean |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 14 of 14 |
| Coverage | All results currently returned for this benchmark |

Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol.

## Metrics

- [Harmonic Mean](<https://www.sota2.com/research/sota/action-recognition-on-ssv2-and-k400.md?metric=Harmonic%20Mean>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Harmonic Mean | Result date | Report paper |
| --- | --- | --- | --- |
| CAST | 77.9 | Not recorded | 68fa74ad1e209131cea6327b |
| OMNIVORE | 77.2 | Not recorded | 68fa74ad1e209131cea6327b |
| Video Swin | 75.8 | Not recorded | 68fa74ad1e209131cea6327b |
| VideoMAE | 75.8 | Not recorded | 68fa74ad1e209131cea6327b |
| ST-Adapter | 75.5 | Not recorded | 68fa74ad1e209131cea6327b |
| AIM | 75.4 | Not recorded | 68fa74ad1e209131cea6327b |
| BEVT | 75.3 | Not recorded | 68fa74ad1e209131cea6327b |
| MTV-HR | 74.8 | Not recorded | 68fa74ad1e209131cea6327b |
| MFormer | 73.7 | Not recorded | 68fa74ad1e209131cea6327b |
| MVIT | 73.4 | Not recorded | 68fa74ad1e209131cea6327b |
| ViViT FE | 73 | Not recorded | 68fa74ad1e209131cea6327b |
| EVL | 71.2 | Not recorded | 68fa74ad1e209131cea6327b |
| TimeSformer | 70.4 | Not recorded | 68fa74ad1e209131cea6327b |
| CLIP | 59.5 | Not recorded | 68fa74ad1e209131cea6327b |

### CAST

Harmonic Mean: 77.9. Result ID: 699353ce3a6c31484915365e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74ad1e209131cea6327b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74ad1e209131cea6327b.md>)

[Implementation](<https://github.com/KHU-VLL/CAST>)

Recorded evaluation configuration: {"GFLOPs/View":"391"}. This record does not assert independent reproduction.

### OMNIVORE

Harmonic Mean: 77.2. Result ID: 699353ce3a6c31484915365c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74ad1e209131cea6327b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6c2d1e209131cea328ef.md>)

[Implementation](<https://facebookresearch.github.io/omnivore>)

Recorded evaluation configuration: {"GFLOPs/View":null}. This record does not assert independent reproduction.

### Video Swin

Harmonic Mean: 75.8. Result ID: 699353ce3a6c314849153659.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74ad1e209131cea6327b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a391e209131cea26be4.md>)

[Implementation](<https://github.com/SwinTransformer/Video-Swin-Transformer>)

Recorded evaluation configuration: {"GFLOPs/View":"282"}. This record does not assert independent reproduction.

### VideoMAE

Harmonic Mean: 75.8. Result ID: 699353ce3a6c31484915365b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74ad1e209131cea6327b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6da81e209131cea3b2fd.md>)

[Implementation](<https://github.com/MCG-NJU/VideoMAE>)

Recorded evaluation configuration: {"GFLOPs/View":"180"}. This record does not assert independent reproduction.

### ST-Adapter

Harmonic Mean: 75.5. Result ID: 699353ce3a6c314849153653.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74ad1e209131cea6327b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6d9f1e209131cea3aeec.md>)

[Implementation](<https://github.com/linziyi96/st-adapter>)

Recorded evaluation configuration: {"GFLOPs/View":"607"}. This record does not assert independent reproduction.

### AIM

Harmonic Mean: 75.4. Result ID: 699353ce3a6c314849153654.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74ad1e209131cea6327b.md>)

Recorded evaluation configuration: {"GFLOPs/View":"404"}. This record does not assert independent reproduction.

### BEVT

Harmonic Mean: 75.3. Result ID: 699353ce3a6c31484915365a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74ad1e209131cea6327b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6bf51e209131cea3138e.md>)

[Implementation](<https://github.com/xyzforever/BEVT>)

Recorded evaluation configuration: {"GFLOPs/View":"282"}. This record does not assert independent reproduction.

### MTV-HR

Harmonic Mean: 74.8. Result ID: 699353ce3a6c31484915365d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74ad1e209131cea6327b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6ca11e209131cea35110.md>)

[Implementation](<https://github.com/google-research/scenic>)

Recorded evaluation configuration: {"GFLOPs/View":"930"}. This record does not assert independent reproduction.

### MFormer

Harmonic Mean: 73.7. Result ID: 699353ce3a6c314849153658.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74ad1e209131cea6327b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6b191e209131cea2c2a8.md>)

[Implementation](<https://github.com/facebookresearch/Motionformer>)

Recorded evaluation configuration: {"GFLOPs/View":"1185"}. This record does not assert independent reproduction.

### MVIT

Harmonic Mean: 73.4. Result ID: 699353ce3a6c314849153657.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74ad1e209131cea6327b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa69cb1e209131cea2412c.md>)

[Implementation](<https://github.com/facebookresearch/SlowFast>)

Recorded evaluation configuration: {"GFLOPs/View":"170"}. This record does not assert independent reproduction.

### ViViT FE

Harmonic Mean: 73. Result ID: 699353ce3a6c314849153655.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74ad1e209131cea6327b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6b2d1e209131cea2c9cc.md>)

[Implementation](<https://github.com/google-research/scenic>)

Recorded evaluation configuration: {"GFLOPs/View":"990"}. This record does not assert independent reproduction.

### EVL

Harmonic Mean: 71.2. Result ID: 699353ce3a6c314849153652.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74ad1e209131cea6327b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6d2d1e209131cea38153.md>)

[Implementation](<https://github.com/OpenGVLab/efficient-video-recognition>)

Recorded evaluation configuration: {"GFLOPs/View":"592"}. This record does not assert independent reproduction.

### TimeSformer

Harmonic Mean: 70.4. Result ID: 699353ce3a6c314849153656.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74ad1e209131cea6327b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a1d1e209131cea25fbb.md>)

[Implementation](<https://github.com/facebookresearch/TimeSformer>)

Recorded evaluation configuration: {"GFLOPs/View":"2380"}. This record does not assert independent reproduction.

### CLIP

Harmonic Mean: 59.5. Result ID: 699353ce3a6c314849153651.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74ad1e209131cea6327b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa69581e209131cea21ad0.md>)

[Implementation](<https://github.com/OpenAI/CLIP>)

Recorded evaluation configuration: {"GFLOPs/View":"140"}. This record does not assert independent reproduction.

Page 1 of 1. This is an index page, not the full collection.