# Action Recognition on LT-NTU IF = 100 120 (cross-subject)

Source: [Action Recognition on LT-NTU IF = 100 120 (cross-subject)](<https://www.sota2.com/research/sota/action-recognition-on-lt-ntu-if-100-120-cross-subject>)

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

Data updated: 2026-02-26T05:54:38.611Z

Evaluation of video action recognition performance on the LT-NTU IF=100 120 dataset using the cross-subject protocol, reporting Top-1 Accuracy broken down by action frequency (Many, Medium, Few).

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) | LT-NTU IF = 100 120 (cross-subject) |
| Metric | Overall Top-1 Acc |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 17 of 17 |
| 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

- [Overall Top-1 Acc](<https://www.sota2.com/research/sota/action-recognition-on-lt-ntu-if-100-120-cross-subject.md?metric=Overall%20Top-1%20Acc>): Higher is better; unit: PERCENT.

- [Top-1 Acc (Many)](<https://www.sota2.com/research/sota/action-recognition-on-lt-ntu-if-100-120-cross-subject.md?metric=Top-1%20Acc%20(Many)>): Higher is better; unit: PERCENT.

- [Top-1 Acc (Medium)](<https://www.sota2.com/research/sota/action-recognition-on-lt-ntu-if-100-120-cross-subject.md?metric=Top-1%20Acc%20(Medium)>): Higher is better; unit: PERCENT.

- [Top-1 Acc (Few)](<https://www.sota2.com/research/sota/action-recognition-on-lt-ntu-if-100-120-cross-subject.md?metric=Top-1%20Acc%20(Few)>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Overall Top-1 Acc | Result date | Report paper |
| --- | --- | --- | --- |
| Shap-Mix | 73 | Not recorded | 68fa74231e209131cea5fe66 |
| GLMC | 71.5 | Not recorded | 68fa74231e209131cea5fe66 |
| BS+Max Norm | 70.3 | Not recorded | 68fa74231e209131cea5fe66 |
| Balanced Softmax (BS) | 69.6 | Not recorded | 68fa74231e209131cea5fe66 |
| Focal Loss | 69.4 | Not recorded | 68fa74231e209131cea5fe66 |
| Remix-DRW | 69.3 | Not recorded | 68fa74231e209131cea5fe66 |
| PaCo | 67.9 | Not recorded | 68fa74231e209131cea5fe66 |
| IB Loss | 67.8 | Not recorded | 68fa74231e209131cea5fe66 |
| Mixup | 66.9 | Not recorded | 68fa74231e209131cea5fe66 |
| BCL | 66.9 | Not recorded | 68fa74231e209131cea5fe66 |
| LDAM-DRW | 66.8 | Not recorded | 68fa74231e209131cea5fe66 |
| FSA | 66.7 | Not recorded | 68fa74231e209131cea5fe66 |
| BRL | 66.3 | Not recorded | 68fa74231e209131cea5fe66 |
| RIDE (3 experts) | 65.2 | Not recorded | 68fa74231e209131cea5fe66 |
| Cross-Entropy Loss | 64.2 | Not recorded | 68fa74231e209131cea5fe66 |
| CB Loss | 63.2 | Not recorded | 68fa74231e209131cea5fe66 |
| ROS | 61 | Not recorded | 68fa74231e209131cea5fe66 |

### Shap-Mix

Overall Top-1 Acc: 73. Result ID: 6993205e3a6c31484913e6be.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74231e209131cea5fe66.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74231e209131cea5fe66.md>)

[Implementation](<https://jhang2020.github.io/Projects/Shap-Mix/Shap-Mix.html>)

Recorded evaluation configuration: {"Category":"Augmentation","Input Stream":"Single Joint"}. This record does not assert independent reproduction.

### GLMC

Overall Top-1 Acc: 71.5. Result ID: 6993205e3a6c31484913e6b2.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74231e209131cea5fe66.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f581e209131cea45243.md>)

[Implementation](<https://github.com/ynu-yangpeng/GLMC>)

Recorded evaluation configuration: {"Category":"Augmentation","Input Stream":"Single Joint"}. This record does not assert independent reproduction.

### BS+Max Norm

Overall Top-1 Acc: 70.3. Result ID: 6993205e3a6c31484913e6ba.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74231e209131cea5fe66.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6c1b1e209131cea32418.md>)

[Implementation](<https://github.com/ShadeAlsha/LTR-weight-balancing>)

Recorded evaluation configuration: {"Category":"Decoupling","Input Stream":"Single Joint"}. This record does not assert independent reproduction.

### Balanced Softmax (BS)

Overall Top-1 Acc: 69.6. Result ID: 6993205e3a6c31484913e6b8.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74231e209131cea5fe66.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa68c11e209131cea1e125.md>)

[Implementation](<https://github.com/jiawei-ren/BalancedMetaSoftmax>)

Recorded evaluation configuration: {"Category":"Reweighting & Resampling","Input Stream":"Single Joint"}. This record does not assert independent reproduction.

### Focal Loss

Overall Top-1 Acc: 69.4. Result ID: 6993205e3a6c31484913e6b5.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74231e209131cea5fe66.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa640c1e209131cea02556.md>)

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

Recorded evaluation configuration: {"Category":"Reweighting & Resampling","Input Stream":"Single Joint"}. This record does not assert independent reproduction.

### Remix-DRW

Overall Top-1 Acc: 69.3. Result ID: 6993205e3a6c31484913e6b0.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74231e209131cea5fe66.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa68c41e209131cea1e3de.md>)

Recorded evaluation configuration: {"Category":"Augmentation","Input Stream":"Single Joint"}. This record does not assert independent reproduction.

### PaCo

Overall Top-1 Acc: 67.9. Result ID: 6993205e3a6c31484913e6bb.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74231e209131cea5fe66.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a8d1e209131cea28b82.md>)

[Implementation](<https://github.com/dvlab-research/Parametric-Contrastive-Learning>)

Recorded evaluation configuration: {"Category":"Contrastive Learning","Input Stream":"Single Joint"}. This record does not assert independent reproduction.

### IB Loss

Overall Top-1 Acc: 67.8. Result ID: 6993205e3a6c31484913e6b9.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74231e209131cea5fe66.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6af71e209131cea2b6c4.md>)

[Implementation](<https://github.com/pseulki/IB-Loss>)

Recorded evaluation configuration: {"Category":"Reweighting & Resampling","Input Stream":"Single Joint"}. This record does not assert independent reproduction.

### Mixup

Overall Top-1 Acc: 66.9. Result ID: 6993205e3a6c31484913e6af.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74231e209131cea5fe66.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa644e1e209131cea03e44.md>)

[Implementation](<https://github.com/facebookresearch/mixup-cifar10>)

Recorded evaluation configuration: {"Category":"Augmentation","Input Stream":"Single Joint"}. This record does not assert independent reproduction.

### BCL

Overall Top-1 Acc: 66.9. Result ID: 6993205e3a6c31484913e6bc.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74231e209131cea5fe66.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6d6c1e209131cea39c49.md>)

Recorded evaluation configuration: {"Category":"Contrastive Learning","Input Stream":"Single Joint"}. This record does not assert independent reproduction.

### LDAM-DRW

Overall Top-1 Acc: 66.8. Result ID: 6993205e3a6c31484913e6b7.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74231e209131cea5fe66.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa66901e209131cea10e0f.md>)

[Implementation](<https://github.com/kaidic/LDAM-DRW>)

Recorded evaluation configuration: {"Category":"Reweighting & Resampling","Input Stream":"Single Joint"}. This record does not assert independent reproduction.

### FSA

Overall Top-1 Acc: 66.7. Result ID: 6993205e3a6c31484913e6b1.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74231e209131cea5fe66.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa68251e209131cea1a42f.md>)

Recorded evaluation configuration: {"Category":"Augmentation","Input Stream":"Single Joint"}. This record does not assert independent reproduction.

### BRL

Overall Top-1 Acc: 66.3. Result ID: 6993205e3a6c31484913e6b3.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74231e209131cea5fe66.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa770b1e209131cea70afa.md>)

Recorded evaluation configuration: {"Category":"Augmentation","Input Stream":"Single Joint"}. This record does not assert independent reproduction.

### RIDE (3 experts)

Overall Top-1 Acc: 65.2. Result ID: 6993205e3a6c31484913e6bd.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74231e209131cea5fe66.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6c7b1e209131cea3436b.md>)

[Implementation](<https://github.com/frank-xwang/RIDE-LongTailRecognition>)

Recorded evaluation configuration: {"Category":"Ensemble","Input Stream":"Single Joint"}. This record does not assert independent reproduction.

### Cross-Entropy Loss

Overall Top-1 Acc: 64.2. Result ID: 6993205e3a6c31484913e6ae.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74231e209131cea5fe66.md>)

Recorded evaluation configuration: {"Category":"-","Input Stream":"Single Joint"}. This record does not assert independent reproduction.

### CB Loss

Overall Top-1 Acc: 63.2. Result ID: 6993205e3a6c31484913e6b6.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74231e209131cea5fe66.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa654f1e209131cea0988a.md>)

[Implementation](<https://github.com/richardaecn/class-balanced-loss>)

Recorded evaluation configuration: {"Category":"Reweighting & Resampling","Input Stream":"Single Joint"}. This record does not assert independent reproduction.

### ROS

Overall Top-1 Acc: 61. Result ID: 6993205e3a6c31484913e6b4.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74231e209131cea5fe66.md>)

Recorded evaluation configuration: {"Category":"Reweighting & Resampling","Input Stream":"Single Joint"}. This record does not assert independent reproduction.

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