| Benchmark | Model | Rank | Results |
|---|---|---|---|
| semantic-segmentation-on-pascal-voc-2012-val | DeepLabv3 (ImageNet+300M) | #17 | mIoU: 76.5% |
| human-pose-keypoint-estimation-on-coco-test-dev | Faster R-CNN (ImageNet+300M) | #95 | AP: 64.4AP50: 85.7AP75: 70.7 |
| image-classification-on-imagenet | ResNet-101 (JFT-300M Finetuning) | #807 | Top 1 Accuracy: 79.2% |
| object-detection-on-coco-test-dev | Faster R-CNN (ImageNet+300M) | – | box mAP: 37.4AP50: 58AP75: 40.1APS: 17.5APM: 41.1APL: 51.2 |