# Video Semantic Segmentation on CaDIS

Source: [Video Semantic Segmentation on CaDIS](<https://www.sota2.com/research/sota/video-semantic-segmentation-on-cadis>)

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

Data updated: 2026-02-26T05:42:50.487Z

Evaluation of Video Semantic Segmentation performance on the CaDIS dataset, reporting metrics such as Mean IoU, Mean Pixel Accuracy, and per-class F1 scores for specific surgical components (e.g., Hand, Eye Retractors, Surgical Tape).

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 | Video Semantic Segmentation |
| Dataset (including any stated split) | CaDIS |
| Metric | F1 (Hand) |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 3 of 3 |
| 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

- [F1 (Hand)](<https://www.sota2.com/research/sota/video-semantic-segmentation-on-cadis.md?metric=F1%20(Hand)>): Higher is better; unit: PERCENT.

- [F1 (Eye Retractors)](<https://www.sota2.com/research/sota/video-semantic-segmentation-on-cadis.md?metric=F1%20(Eye%20Retractors)>): Higher is better; unit: PERCENT.

- [F1 (Surgical Tape)](<https://www.sota2.com/research/sota/video-semantic-segmentation-on-cadis.md?metric=F1%20(Surgical%20Tape)>): Higher is better; unit: PERCENT.

- [F1 (Instrument)](<https://www.sota2.com/research/sota/video-semantic-segmentation-on-cadis.md?metric=F1%20(Instrument)>): Higher is better; unit: PERCENT.

- [F1 (Pupil)](<https://www.sota2.com/research/sota/video-semantic-segmentation-on-cadis.md?metric=F1%20(Pupil)>): Higher is better; unit: PERCENT.

- [F1 (Skin)](<https://www.sota2.com/research/sota/video-semantic-segmentation-on-cadis.md?metric=F1%20(Skin)>): Higher is better; unit: PERCENT.

- [F1 (Iris)](<https://www.sota2.com/research/sota/video-semantic-segmentation-on-cadis.md?metric=F1%20(Iris)>): Higher is better; unit: PERCENT.

- [F1 (Cornea)](<https://www.sota2.com/research/sota/video-semantic-segmentation-on-cadis.md?metric=F1%20(Cornea)>): Higher is better; unit: PERCENT.

- [Mean F1](<https://www.sota2.com/research/sota/video-semantic-segmentation-on-cadis.md?metric=Mean%20F1>): Higher is better; unit: PERCENT.

- [Mean IoU](<https://www.sota2.com/research/sota/video-semantic-segmentation-on-cadis.md?metric=Mean%20IoU>): Higher is better; unit: PERCENT.

- [Mean Pixel Accuracy](<https://www.sota2.com/research/sota/video-semantic-segmentation-on-cadis.md?metric=Mean%20Pixel%20Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | F1 (Hand) | Result date | Report paper |
| --- | --- | --- | --- |
| TCNN | 95.29 | Not recorded | 68fa6b7c1e209131cea2e7f2 |
| Image Predictor | 94.59 | Not recorded | 68fa6b7c1e209131cea2e7f2 |
| Video Predictor | 92.73 | Not recorded | 68fa6b7c1e209131cea2e7f2 |

### TCNN

F1 (Hand): 95.29. Result ID: 698e8f9dffc3901de09ea947.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6b7c1e209131cea2e7f2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6b7c1e209131cea2e7f2.md>)

Recorded evaluation configuration: {"Backbone":"ResNet-50","Additional Supervision":"Autoencoder Network"}. This record does not assert independent reproduction.

### Image Predictor

F1 (Hand): 94.59. Result ID: 698e8f9dffc3901de09ea945.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6b7c1e209131cea2e7f2.md>)

Recorded evaluation configuration: {"Backbone":"ResNet-50","Base Model":"DeepLabv3+"}. This record does not assert independent reproduction.

### Video Predictor

F1 (Hand): 92.73. Result ID: 698e8f9dffc3901de09ea946.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6b7c1e209131cea2e7f2.md>)

Recorded evaluation configuration: {"Backbone":"ResNet-50","Base Model":"DeepLabv3+","Temporal Module":"3x3 ConvLSTM"}. This record does not assert independent reproduction.

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