NR cnniqa, dbcnn, hyperiqa, wadiqam_nr Aesthetic IQA nima, nima-vgg16-avaNote Due to optimized training process, performance of some retrained approaches may be different with original paper. Results of all retrained models by ours are normalized to [0, 1] and change to higher better for ...
"ARNIQA": f"{base_url}/ARNIQA.pth", "live": f"{base_url}/regressor_live.pth", "csiq": f"{base_url}/regressor_csiq.pth", "tid": f"{base_url}/regressor_tid2013.pth", "kadid": f"{base_url}/regressor_kadid10k.pth", "koniq": f"{base_url}/regressor_koniq10k.pth", "clive...