chelt

没有复现IFAN,但如果你直接用IFAN trained by 8bit images的模型在16bit图像跑,效果是这样的: IFAN Overall: PSNR 26.023693 SSIM 0.805778 MAE 0.037130 LPIPS 0.805778 Indoor: PSNR 28.699427 SSIM 0.870385 MAE 0.024749 LPIPS 0.024749 Outdoor: PSNR 23.485176 SSIM 0.744484 MAE...

Python 3.8.10 >>> torch.__version__ '2.0.0+cu117' 我的环境可以正常运行他运动模糊实验的结果,pip list后是这样的输出: absl-py 1.4.0 addict 2.4.0 antlr4-python3-runtime 4.9.3 beautifulsoup4 4.12.2 blinker 1.6.2 cachetools 5.3.0 certifi 2022.12.7 charset-normalizer 3.1.0 click 8.1.3 cmake 3.26.1 contourpy 1.0.7 cycler...