Commit e45d920a authored by comfyanonymous's avatar comfyanonymous
Browse files

Don't resize clip vision image when the size is already good.

parent 13e6d536
...@@ -19,8 +19,10 @@ class Output: ...@@ -19,8 +19,10 @@ class Output:
def clip_preprocess(image, size=224): def clip_preprocess(image, size=224):
mean = torch.tensor([ 0.48145466,0.4578275,0.40821073], device=image.device, dtype=image.dtype) mean = torch.tensor([ 0.48145466,0.4578275,0.40821073], device=image.device, dtype=image.dtype)
std = torch.tensor([0.26862954,0.26130258,0.27577711], device=image.device, dtype=image.dtype) std = torch.tensor([0.26862954,0.26130258,0.27577711], device=image.device, dtype=image.dtype)
scale = (size / min(image.shape[1], image.shape[2])) image = image.movedim(-1, 1)
image = torch.nn.functional.interpolate(image.movedim(-1, 1), size=(round(scale * image.shape[1]), round(scale * image.shape[2])), mode="bicubic", antialias=True) if not (image.shape[2] == size and image.shape[3] == size):
scale = (size / min(image.shape[2], image.shape[3]))
image = torch.nn.functional.interpolate(image, size=(round(scale * image.shape[2]), round(scale * image.shape[3])), mode="bicubic", antialias=True)
h = (image.shape[2] - size)//2 h = (image.shape[2] - size)//2
w = (image.shape[3] - size)//2 w = (image.shape[3] - size)//2
image = image[:,:,h:h+size,w:w+size] image = image[:,:,h:h+size,w:w+size]
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment