Wrong preprocessing

[JunkyByte]

Training preprocessing is missing during inference where images are only set to [0, 1] range but not normalized using mean / std.

Inference code:

ViTPose_pytorch/inference.py

Lines 46 to 48 in 1bd3cc3

	img_tensor = transforms.Compose (
	[transforms.Resize((img_size[1], img_size[0])),
	transforms.ToTensor()]

Training code:

ViTPose_pytorch/datasets/COCO.py

Lines 119 to 122 in 1bd3cc3

	self.transform = transforms.Compose([
	transforms.ToTensor(),
	transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]),
	])

Reference from issue that was created in my fork
JunkyByte/easy_ViTPose#12