Guanying Chen
Guanying Chen
Hi, you might try ``` CUDA_VISIBLE_DEVICES=0 python main_stage2.py --in_img_num 32 --retrain data/logdir/path/to/checkpointDirOfLCNet/checkpoint20.pth.tar --retrain_s2 /path/to/your/half-trained-stage2-mode.pth.tar ``` You might need to adjust the initial learning rate as well.
Thanks for the information, we have fixed it.
Hi, those HDR images are results of the Kalantari 13's method.
We do not pre-align when testing on our dataset because we are evaluating the robustness to global motion. I think global image alignment is helpful as it reduce the difficulty...
Hi, I think you can just specify the checkpoint path as like ``` python main.py --gpu_ids 0 --model hdr2E_flow2s_model --fnet_checkp /path/to/flow_net.pth --mnet_checkp /path/to/weight_net.pth \ --mnet2_checkp /path/to/mnet2_checkp14.pt ```
Hi, may I ask the following questions? 1. What is the configuration for the three-stop exposure? Are you using {EV-3, EV+0, EV+3, EV-3, EV+0...} or {EV-3, EV+0, EV+3, EV+0, EV-3...}....
Hi, I understand you question now. When using 3-exposure-2stop or 3-exposure-1stop videos, the results are reasonable. But if tested with 3-exposure-3stop video, the results are flickering. This observation is interesting,...
Ok, you could send me the dataset (e.g., a google drive link to [email protected]). I will have a check when I have time. Thanks.
Hi, it seems that I haven't received the link.
Hi, I have tested the code and the results are similar to yours. I think there is a bug in the code, which might be the thresholds used for determining...