Closed Williamlizl closed 3 years ago
Regarding blur synthesis, you can refer to #13. GOPRO_Large_all means all the frames used to generate blur. You can take a look at the description here.
Regarding blur synthesis, you can refer to #13. GOPRO_Large_all means all the frames used to generate blur. You can take a look at the description here. `img1 = image.load(filename1) img2 = image.load(filename2)
signal1 = torch.pow(img1, 2.2) signal2 = torch.pow(img2, 2.2)
signal_blur = (signal1 + signal2)/2 img_blur = torch.pow(signal_blur, 1/2.2)` image.load? or can i use other ways to generate blur images?https://github.com/codebox/image_augmentor https://github.com/aleju/imgaug
What is mentioned in #13 is a torch (Lua) version pseudocode as this repository builds on it. You can use whichever tool you prefer.
But I'd recommend you to refer to our updated dataset, REDS for advanced blur synthesis.
every sharp photo matches the blurred photo? or just sharp the original photo
I'm not sure if I understand your question. In GOPRO_Large, I provide the blurry and the sharp image pairs that are temporally center-aligned. In GOPRO_Large_all, I provide all the sharp frames I used to generate blurry images.
I mean if I have one original image, then I generate the paired blurry image. But for the sharp image, I also need to generate it by the original image?
A sharp image is the original image itself.
You need multiple sharp images to synthesize a blurry image.
A sharp image is the original image itself.
You need multiple sharp images to synthesize a blurry image.
Thank you, I will try it.
Regarding blur synthesis, you can refer to #13. GOPRO_Large_all means all the frames used to generate blur. You can take a look at the description here.