You are not limited to the LOL dataset. You can place any images in source/low/. The algorithm will process whatever files it finds there: Why 64? 64 = 128/2: one exposure stop below the perceptually ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results