Image-Adaptive-3DLUT
Image-Adaptive-3DLUT copied to clipboard
HuiZeng
Learning Image-adaptive 3D Lookup Tables for High Performance Photo Enhancement in Real-time
Hi, in your paper, you said "It should be noted that we only learn the color/tone enhancement part of the imaging pipeline rather than learning the entire pipeline from raw...
Hello, Thank you for your contribution. I see that you have shared two models(LUTs.pth, classifier.pth) for sRGB paired training approach [here](https://github.com/HuiZeng/Image-Adaptive-3DLUT/tree/master/pretrained_models/sRGB). Can you please share the following information: - Which...
Hi, First, some basic information about my setup which may be relevant. I am running Pyton 3.8.1, which was installed as part of Anaconda. I am using the Spyder IDE....
In [datasets.py](https://github.com/HuiZeng/Image-Adaptive-3DLUT/blob/master/datasets.py#L133) I can see the following: ``` img_input = cv2.imread(self.set1_input_files[index % len(self.set1_input_files)],-1) img_exptC = Image.open(self.set1_expert_files[index % len(self.set1_expert_files)]) ``` Here img_input is in XYZ while img_exptC is in RGB. But...
VS 2019 windows 10 pytorch 1.6 cuda 10.1 install  i do not to do about this error ,can you help me ,thanks
I noticed the CPU implementation of trilinear forward and backward functions only support batch_size == 1. When we use the functions with batch_size > 2, the first sample is only...
hello, how can i solve this problem? I looked it up online and said it was memory or Pointers?
你好，阅读了您的代码后，有个想法希望沟通一下： 从训练的时候看lut表关系大致表述为： result = w1 * lut1(img) + ...+ wn*lutn(img) ------1 从demo_val中看后面的表述关系大致为： blendlut = w1*lut1+...+wn*lutn result = blendlut(img) -------2 假如上面的1和2等价的话， 那么是否就是说 如果训练足够多，应该就能用一个lut表来表征呢？ 另外一个问题是：如何评价训练结果，即什么时候可以中止训练？ 期待您的答复 谢谢
In some situation, training code raises a device error: ``` cuda error: an illegal memory access was encountered ``` After debug I found that the main reason is `Tensor` used...
