cchen156

Hi Yuanming, Thank you for your update! For dataset, you are free to use our data: https://github.com/cchen156/Learning-to-See-in-the-Dark Thank you!

Can you paste one example here? Both input and output of the network. Did you check whether you have the pertained model ready?

Please try the Baidu drive link.

Please check #1 #2

I think you can apply WB before network. It should work.

You can copy the network and train with your low-light, ground-truth pairs. I am not sure how you darken it using PS.

You need raw data. like DNG, ARW, etc. Not PNG or JPG

I used rawpy to process the raw images to sRGB images. And then use these images as input to train the network for comparisons.

I trained the model using our data. I saved the raw data into 16-bit png files and trained the model using their code.

What I did is: 1. subtract black level 2. pack Bayer raw data into RGB channels. The green pixel is the average of two green pixels in each 2*2 block....