About [CVPR 2024] The official implementation of paper " Light the Night: A Multi-Condition Diffusion Framework for Unpaired Low-Light Enhancement in Autonomous Driving"
Apache License 2.0
55
stars
0
forks
source link
Consulting for the loss function in LiDAR and Distribution Reward Modeling #2
When I read the paper, I find that the loss in LiDAR and Distribution Reward Modeling section is confusing. Can you explain more about the Reinforcement Learning (RL) training strategy utilized for the loss in the paper?
Hi authors,
Thanks for your nice work!
When I read the paper, I find that the loss in LiDAR and Distribution Reward Modeling section is confusing. Can you explain more about the Reinforcement Learning (RL) training strategy utilized for the loss in the paper?
Hope to hear from you soon!