# Action Recognition on NTU RGB+D X-Sub protocol 120 (Hard)

Source: [Action Recognition on NTU RGB+D X-Sub protocol 120 (Hard)](<https://www.sota2.com/research/sota/action-recognition-on-ntu-rgb-plus-d-x-sub-protocol-120-hard>)

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

Data updated: 2026-07-02T08:58:07.240Z

Evaluation of video action recognition performance using the NTU RGB+D 120 dataset specifically under the Cross-Subject (X-Sub) protocol, focusing on the 'Hard' difficulty subset, measured by Accuracy.

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) | NTU RGB+D X-Sub protocol 120 (Hard) |
| Metric | Accuracy (%) |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 15 of 15 |
| 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

- [Accuracy (%)](<https://www.sota2.com/research/sota/action-recognition-on-ntu-rgb-plus-d-x-sub-protocol-120-hard.md?metric=Accuracy%20(%25)>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy (%) | Result date | Report paper |
| --- | --- | --- | --- |
| PartialVisGraph | 72.9 | Not recorded | 6a45f071dc71372d7795a0af |
| PartialVisGraph | 71.5 | Not recorded | 6a45f071dc71372d7795a0af |
| PartialVisGraph | 66.7 | Not recorded | 6a45f071dc71372d7795a0af |
| Hyperformer | 21 | Not recorded | 6a45f071dc71372d7795a0af |
| Hyperformer | 16.6 | Not recorded | 6a45f071dc71372d7795a0af |
| FR-Head | 12.7 | Not recorded | 6a45f071dc71372d7795a0af |
| Hyper-GCN | 12.7 | Not recorded | 6a45f071dc71372d7795a0af |
| HD-GCN | 12.2 | Not recorded | 6a45f071dc71372d7795a0af |
| FR-Head | 10 | Not recorded | 6a45f071dc71372d7795a0af |
| HD-GCN | 9.2 | Not recorded | 6a45f071dc71372d7795a0af |
| Hyper-GCN | 9.1 | Not recorded | 6a45f071dc71372d7795a0af |
| Hyperformer | 4.7 | Not recorded | 6a45f071dc71372d7795a0af |
| FR-Head | 4.3 | Not recorded | 6a45f071dc71372d7795a0af |
| HD-GCN | 2.8 | Not recorded | 6a45f071dc71372d7795a0af |
| Hyper-GCN | 1.6 | Not recorded | 6a45f071dc71372d7795a0af |

### PartialVisGraph

Accuracy (%): 72.9. Result ID: 6a46281e60b88bb339859192.

[Reporting paper](<https://www.sota2.com/research/paper/6a45f071dc71372d7795a0af.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a45f071dc71372d7795a0af.md>)

Recorded evaluation configuration: {"FoV centre":"Head (3)"}. This record does not assert independent reproduction.

### PartialVisGraph

Accuracy (%): 71.5. Result ID: 6a46281e60b88bb339859197.

[Reporting paper](<https://www.sota2.com/research/paper/6a45f071dc71372d7795a0af.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a45f071dc71372d7795a0af.md>)

Recorded evaluation configuration: {"FoV centre":"Spine (20)"}. This record does not assert independent reproduction.

### PartialVisGraph

Accuracy (%): 66.7. Result ID: 6a46281e60b88bb33985919c.

[Reporting paper](<https://www.sota2.com/research/paper/6a45f071dc71372d7795a0af.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a45f071dc71372d7795a0af.md>)

Recorded evaluation configuration: {"FoV centre":"Base of the spine (0)"}. This record does not assert independent reproduction.

### Hyperformer

Accuracy (%): 21. Result ID: 6a46281e60b88bb339859191.

[Reporting paper](<https://www.sota2.com/research/paper/6a45f071dc71372d7795a0af.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6ecc1e209131cea42362.md>)

[Implementation](<https://www.sota2.com/Hyperformer%20Github%20Repository>)

Recorded evaluation configuration: {"FoV centre":"Head (3)"}. This record does not assert independent reproduction.

### Hyperformer

Accuracy (%): 16.6. Result ID: 6a46281e60b88bb339859196.

[Reporting paper](<https://www.sota2.com/research/paper/6a45f071dc71372d7795a0af.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6ecc1e209131cea42362.md>)

[Implementation](<https://www.sota2.com/Hyperformer%20Github%20Repository>)

Recorded evaluation configuration: {"FoV centre":"Spine (20)"}. This record does not assert independent reproduction.

### FR-Head

Accuracy (%): 12.7. Result ID: 6a46281e60b88bb33985918e.

[Reporting paper](<https://www.sota2.com/research/paper/6a45f071dc71372d7795a0af.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6edf1e209131cea428c6.md>)

[Implementation](<https://github.com/zhysora/FR-Head>)

Recorded evaluation configuration: {"FoV centre":"Head (3)"}. This record does not assert independent reproduction.

### Hyper-GCN

Accuracy (%): 12.7. Result ID: 6a46281e60b88bb339859190.

[Reporting paper](<https://www.sota2.com/research/paper/6a45f071dc71372d7795a0af.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a611e209131cea82bb8.md>)

Recorded evaluation configuration: {"FoV centre":"Head (3)"}. This record does not assert independent reproduction.

### HD-GCN

Accuracy (%): 12.2. Result ID: 6a46281e60b88bb33985918f.

[Reporting paper](<https://www.sota2.com/research/paper/6a45f071dc71372d7795a0af.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa701e1e209131cea492aa.md>)

[Implementation](<https://github.com/Jho-Yonsei/HD-GCN>)

Recorded evaluation configuration: {"FoV centre":"Head (3)"}. This record does not assert independent reproduction.

### FR-Head

Accuracy (%): 10. Result ID: 6a46281e60b88bb339859193.

[Reporting paper](<https://www.sota2.com/research/paper/6a45f071dc71372d7795a0af.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6edf1e209131cea428c6.md>)

[Implementation](<https://github.com/zhysora/FR-Head>)

Recorded evaluation configuration: {"FoV centre":"Spine (20)"}. This record does not assert independent reproduction.

### HD-GCN

Accuracy (%): 9.2. Result ID: 6a46281e60b88bb339859194.

[Reporting paper](<https://www.sota2.com/research/paper/6a45f071dc71372d7795a0af.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa701e1e209131cea492aa.md>)

[Implementation](<https://github.com/Jho-Yonsei/HD-GCN>)

Recorded evaluation configuration: {"FoV centre":"Spine (20)"}. This record does not assert independent reproduction.

### Hyper-GCN

Accuracy (%): 9.1. Result ID: 6a46281e60b88bb339859195.

[Reporting paper](<https://www.sota2.com/research/paper/6a45f071dc71372d7795a0af.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a611e209131cea82bb8.md>)

Recorded evaluation configuration: {"FoV centre":"Spine (20)"}. This record does not assert independent reproduction.

### Hyperformer

Accuracy (%): 4.7. Result ID: 6a46281e60b88bb33985919b.

[Reporting paper](<https://www.sota2.com/research/paper/6a45f071dc71372d7795a0af.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6ecc1e209131cea42362.md>)

[Implementation](<https://www.sota2.com/Hyperformer%20Github%20Repository>)

Recorded evaluation configuration: {"FoV centre":"Base of the spine (0)"}. This record does not assert independent reproduction.

### FR-Head

Accuracy (%): 4.3. Result ID: 6a46281e60b88bb339859198.

[Reporting paper](<https://www.sota2.com/research/paper/6a45f071dc71372d7795a0af.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6edf1e209131cea428c6.md>)

[Implementation](<https://github.com/zhysora/FR-Head>)

Recorded evaluation configuration: {"FoV centre":"Base of the spine (0)"}. This record does not assert independent reproduction.

### HD-GCN

Accuracy (%): 2.8. Result ID: 6a46281e60b88bb339859199.

[Reporting paper](<https://www.sota2.com/research/paper/6a45f071dc71372d7795a0af.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa701e1e209131cea492aa.md>)

[Implementation](<https://github.com/Jho-Yonsei/HD-GCN>)

Recorded evaluation configuration: {"FoV centre":"Base of the spine (0)"}. This record does not assert independent reproduction.

### Hyper-GCN

Accuracy (%): 1.6. Result ID: 6a46281e60b88bb33985919a.

[Reporting paper](<https://www.sota2.com/research/paper/6a45f071dc71372d7795a0af.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a611e209131cea82bb8.md>)

Recorded evaluation configuration: {"FoV centre":"Base of the spine (0)"}. This record does not assert independent reproduction.

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