Zhenyi Liu

Results 3 issues of Zhenyi Liu

Thanks for sharing such great work, I wonder is it possible to adapt the code to support the detection models from torchvision?

Is mask used in [forward function](https://github.com/wyf0912/ExposureDiffusion/blob/29b125a0ae2cc0ca5f52395c65f4f4e340a20f97/models/losses.py#L33C14-L33C14) a different one than the soft mask used in output function? (seems different) And is there any insight on the the threshold value? Thanks!

When mixed textures (such as the watercolor scene in pbrt-v4-scenes) are used with NSpectralSamples larger than 20, I encounter a CUDA error.: ``` FATAL CUDA error: an illegal memory access...