Object Detection On Coco O
Metrics
Average mAP
Effective Robustness
Results
Performance results of various models on this benchmark
Comparison Table
| Model Name | Average mAP | Effective Robustness |
|---|---|---|
| ssd-single-shot-multibox-detector | 13.6 | 0.36 |
| grit-a-generative-region-to-text-transformer | 42.9 | 15.72 |
| usb-universal-scale-object-detection | - | 1.86 |
| coarse-to-fine-vision-language-pre-training | 33.7 | 11.43 |
| dynamic-head-unifying-object-detection-heads | 19.3 | 0.16 |
| mask-r-cnn | 17.1 | - |
| varifocalnet-an-iou-aware-dense-object | 28.0 | 5.27 |
| nms-strikes-back | 48.5 | 20.15 |
| mask-r-cnn | - | -0.11 |
| gcnet-non-local-networks-meet-squeeze | 26.0 | 4.38 |
| focal-loss-for-dense-object-detection | 16.6 | 0.18 |
| yolov7-trainable-bag-of-freebies-sets-new | 32.0 | 6.42 |
| yolox-exceeding-yolo-series-in-2021 | 30.3 | 7.26 |
| dino-detr-with-improved-denoising-anchor-1 | 42.1 | 15.76 |
| grounded-language-image-pre-training | 48.0 | 24.89 |
| yolox-exceeding-yolo-series-in-2021 | 20.6 | 2.48 |
| you-only-look-at-one-sequence-rethinking | 20.0 | 1.05 |
| dynamic-head-unifying-object-detection-heads | 35.3 | 10.00 |
| yolov3-an-incremental-improvement | 14.8 | -0.37 |
| reppoints-v2-verification-meets-regression | 24.9 | 2.7 |
| vision-transformer-adapter-for-dense | 34.25 | 7.79 |
| internimage-exploring-large-scale-vision | 37.0 | 11.72 |
| grounded-language-image-pre-training | 29.1 | 8.11 |
| generalized-focal-loss-v2-learning-reliable | 25.1 | 2.6 |
| eva-exploring-the-limits-of-masked-visual | 57.8 | 28.86 |
| hybrid-task-cascade-for-instance-segmentation | 19.1 | 0.08 |
| usb-universal-scale-object-detection | 24.8 | - |
| queryinst-parallelly-supervised-mask-query | 33.2 | 8.26 |
| deformable-detr-deformable-transformers-for-1 | 18.5 | -1.49 |
| yolov6-a-single-stage-object-detection | 32.5 | 6.73 |
| fcos-fully-convolutional-one-stage-object | 16.7 | 0.25 |
| cascade-r-cnn-high-quality-object-detection | 18.2 | 0.02 |
| improved-multiscale-vision-transformers-for | 30.9 | 5.62 |
| efficientdet-scalable-and-efficient-object | 28.5 | 5.44 |
| probabilistic-two-stage-detection | 29.5 | 4.29 |
| robust-and-accurate-object-detection-via | 30.8 | 7.34 |
| yolov4-optimal-speed-and-accuracy-of-object | 30.4 | 5.89 |
| end-to-end-object-detection-with-transformers | 17.1 | -1.82 |
| pvtv2-improved-baselines-with-pyramid-vision | 28.2 | 6.85 |
| bridging-the-gap-between-anchor-based-and | 16.8 | -0.91 |
| a-convnet-for-the-2020s | 37.5 | 12.68 |
| cbnetv2-a-composite-backbone-network | 39.0 | 12.36 |
| exploring-plain-vision-transformer-backbones | - | 7.89 |
| exploring-plain-vision-transformer-backbones | 34.3 | - |
| faster-r-cnn-towards-real-time-object | 16.4 | -0.41 |