l am new to image processing.I ran the code given by the author of KinD(the corresponding paper is Kindling the Darkness--A Practical Low-light Image Enhancer), used the author's pre-trained network, and ran the "evaluate.py" file, and then I got the results, but the output images did not correct.
The first image is the low-light image given by the author for testing, and the followings are the outputs of "adjust_i","decom_i_low","decom_r_low","denoise_r" and "kindle".Why could this happen? Thanks for your help!
origin https://i.stack.imgur.com/N2WB1.jpg
adjust_i https://i.stack.imgur.com/VkKho.png
decom_i_low https://i.stack.imgur.com/bT4Yp.png
decom_r_low https://i.stack.imgur.com/rBA3P.png
denoise_r https://i.stack.imgur.com/am4Th.png
kindle https://i.stack.imgur.com/cunF5.png
