Thanks for your remarkable work. We tested the pre-trained model you provided on the REDS dataset and got excellent results, but the performance on our own data (captured with an iPhone) was quite unsatisfactory. May I ask what might be the reason for this? Alternatively, are there any parameter settings we should pay attention to when testing on our own dataset?
Thanks for your remarkable work. We tested the pre-trained model you provided on the REDS dataset and got excellent results, but the performance on our own data (captured with an iPhone) was quite unsatisfactory. May I ask what might be the reason for this? Alternatively, are there any parameter settings we should pay attention to when testing on our own dataset?