Zsugabubus27
commented
4 years ago Hi!
It is because the Extractor expects list of images, however it is called with a single image
Change this:
feature = extr(img)
to this:
feature = extr([img])
Note that this does not affects the behaviour of the whole algorithm since running feature_extractor.py only is just for testing.
I tried to run deep_sort/deep/feature_extractor.py, but there was a problem.
Traceback (most recent call last): File "D:/目标检测各模型/deep_sort_pytorch-master/deep_sort/deep/feature_extractor.py", line 56, in
feature = extr(img)
File "D:/目标检测各模型/deep_sort_pytorch-master/deep_sort/deep/feature_extractor.py", line 44, in call
im_batch = self._preprocess(im_crops)
File "D:/目标检测各模型/deep_sort_pytorch-master/deep_sort/deep/feature_extractor.py", line 39, in _preprocess
im_batch = torch.cat([self.norm(_resize(im, self.size)).unsqueeze(0) for im in im_crops], dim=0).float()
File "D:/目标检测各模型/deep_sort_pytorch-master/deep_sort/deep/feature_extractor.py", line 39, in
im_batch = torch.cat([self.norm(_resize(im, self.size)).unsqueeze(0) for im in imcrops], dim=0).float()
File "F:\softmanage\Anaconda3\lib\site-packages\torchvision\transforms\transforms.py", line 61, in call
img = t(img)
File "F:\softmanage\Anaconda3\lib\site-packages\torchvision\transforms\transforms.py", line 166, in call
return F.normalize(tensor, self.mean, self.std, self.inplace)
File "F:\softmanage\Anaconda3\lib\site-packages\torchvision\transforms\functional.py", line 208, in normalize
tensor.sub(mean).div_(std)
RuntimeError: output with shape [1, 128, 64] doesn't match the broadcast shape [3, 128, 64]
can you help me ? thanks!!!