Improving Augmentation and Evaluation Schemes for Semantic Image Synthesis

Open paper
Benchmark Model Rank Results
image-to-image-translation-on-ade20k-labels-to-photosCC-FPSE-AUG–mIoU: 44Accuracy: 83%FID: 32.6
image-to-image-translation-on-ade20k-labels-to-photosPix2PixHD-AUG–Accuracy: 77.9%FID: 41.5
image-to-image-translation-on-cityscapes-labels-to-photoCC-FPSE-AUG–mIoU: 63.1FID: 52.1Accuracy: 93.5
image-to-image-translation-on-cityscapes-labels-to-photoPix2PixHD-AUG–mIoU: 58FID: 72.7Accuracy: 92.7
image-to-image-translation-on-coco-stuff-labels-to-photosCC-FPSE-AUG–FID: 19.1Accuracy: 71.5mIoU: 42.1
image-to-image-translation-on-coco-stuff-labels-to-photosPix2PixHD-AUG–FID: 54.2Accuracy: 54.1mIoU: 21.9

Rank counts only results with a code link.