# Object Detection on In-house TCT image dataset (test)

Source: [Object Detection on In-house TCT image dataset (test)](<https://www.sota2.com/research/sota/object-detection-on-in-house-tct-image-dataset-test>)

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

Data updated: 2026-02-26T05:23:14.283Z

Evaluation of Object Detection performance on the test split of the In-house TCT image dataset.

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 | Object Detection |
| Dataset (including any stated split) | In-house TCT image dataset (test) |
| Metric | AP \[0.2:0.5\] |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 10 of 10 |
| 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

- [AP \[0.2:0.5\]](<https://www.sota2.com/research/sota/object-detection-on-in-house-tct-image-dataset-test.md?metric=AP%20%5B0.2%3A0.5%5D>): Higher is better; unit: PERCENT.

- [AP (IoU \<= 0.2)](<https://www.sota2.com/research/sota/object-detection-on-in-house-tct-image-dataset-test.md?metric=AP%20(IoU%20%3C%3D%200.2)>): Higher is better; unit: PERCENT.

- [AP @ IoU=0.5](<https://www.sota2.com/research/sota/object-detection-on-in-house-tct-image-dataset-test.md?metric=AP%20%40%20IoU%3D0.5>): Higher is better; unit: PERCENT.

- [AR](<https://www.sota2.com/research/sota/object-detection-on-in-house-tct-image-dataset-test.md?metric=AR>): Higher is better; unit: PERCENT.

## Recorded results

| Method | AP \[0.2:0.5\] | Result date | Report paper |
| --- | --- | --- | --- |
| Gaze-DINO | 68.7 | Not recorded | 68fa73371e209131cea5b3bb |
| DINO | 64.6 | Not recorded | 68fa73371e209131cea5b3bb |
| Gaze-DN-DETR | 57.3 | Not recorded | 68fa73371e209131cea5b3bb |
| DN-DETR | 53.5 | Not recorded | 68fa73371e209131cea5b3bb |
| DETR | 51.1 | Not recorded | 68fa73371e209131cea5b3bb |
| Conditional-DETR | 48.3 | Not recorded | 68fa73371e209131cea5b3bb |
| YOLOv8 | 48.2 | Not recorded | 68fa73371e209131cea5b3bb |
| RetinaNet | 46.6 | Not recorded | 68fa73371e209131cea5b3bb |
| EllipseNet | 44.3 | Not recorded | 68fa73371e209131cea5b3bb |
| Sparse R-CNN | 38.6 | Not recorded | 68fa73371e209131cea5b3bb |

### Gaze-DINO

AP \[0.2:0.5\]: 68.7. Result ID: 6992cc603a6c314849119225.

[Reporting paper](<https://www.sota2.com/research/paper/68fa73371e209131cea5b3bb.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73371e209131cea5b3bb.md>)

Recorded evaluation configuration: {"Backbone":"Swin-L"}. This record does not assert independent reproduction.

### DINO

AP \[0.2:0.5\]: 64.6. Result ID: 6992cc603a6c314849119223.

[Reporting paper](<https://www.sota2.com/research/paper/68fa73371e209131cea5b3bb.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6cf41e209131cea36cf4.md>)

[Implementation](<https://github.com/IDEACVR/DINO>)

Recorded evaluation configuration: {"Backbone":"Swin-L"}. This record does not assert independent reproduction.

### Gaze-DN-DETR

AP \[0.2:0.5\]: 57.3. Result ID: 6992cc603a6c314849119224.

[Reporting paper](<https://www.sota2.com/research/paper/68fa73371e209131cea5b3bb.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73371e209131cea5b3bb.md>)

Recorded evaluation configuration: {"Backbone":"ResNet-50"}. This record does not assert independent reproduction.

### DN-DETR

AP \[0.2:0.5\]: 53.5. Result ID: 6992cc603a6c314849119222.

[Reporting paper](<https://www.sota2.com/research/paper/68fa73371e209131cea5b3bb.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e2e1e209131cea3e653.md>)

[Implementation](<https://github.com/IDEA-Research/DN-DETR>)

Recorded evaluation configuration: {"Backbone":"ResNet-50"}. This record does not assert independent reproduction.

### DETR

AP \[0.2:0.5\]: 51.1. Result ID: 6992cc603a6c314849119220.

[Reporting paper](<https://www.sota2.com/research/paper/68fa73371e209131cea5b3bb.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67ad1e209131cea17bce.md>)

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

Recorded evaluation configuration: {"Backbone":"ResNet-50"}. This record does not assert independent reproduction.

### Conditional-DETR

AP \[0.2:0.5\]: 48.3. Result ID: 6992cc603a6c314849119221.

[Reporting paper](<https://www.sota2.com/research/paper/68fa73371e209131cea5b3bb.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70ba1e209131cea4cdda.md>)

[Implementation](<https://github.com/Atten4Vis/ConditionalDETR>)

Recorded evaluation configuration: {"Backbone":"ResNet-50"}. This record does not assert independent reproduction.

### YOLOv8

AP \[0.2:0.5\]: 48.2. Result ID: 6992cc603a6c31484911921f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa73371e209131cea5b3bb.md>)

Recorded evaluation configuration: {"Backbone":"ResNet-50"}. This record does not assert independent reproduction.

### RetinaNet

AP \[0.2:0.5\]: 46.6. Result ID: 6992cc603a6c31484911921d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa73371e209131cea5b3bb.md>)

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

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

Recorded evaluation configuration: {"Backbone":"ResNet-50"}. This record does not assert independent reproduction.

### EllipseNet

AP \[0.2:0.5\]: 44.3. Result ID: 6992cc603a6c31484911921c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa73371e209131cea5b3bb.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6ad51e209131cea2aa35.md>)

[Implementation](<https://git.openi.org.cn/capepoint/EllipseNet>)

Recorded evaluation configuration: {"Backbone":"ResNet-50"}. This record does not assert independent reproduction.

### Sparse R-CNN

AP \[0.2:0.5\]: 38.6. Result ID: 6992cc603a6c31484911921e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa73371e209131cea5b3bb.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa69cd1e209131cea24256.md>)

[Implementation](<https://github.com/PeizeSun/SparseR-CNN>)

Recorded evaluation configuration: {"Backbone":"ResNet-50"}. This record does not assert independent reproduction.

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