Super-six-java
commented
1 year ago convert_vedio()怎么使用图片序列啊，我这边一直报错
0%| | 0/208 [00:00<?, ?it/s]Traceback (most recent call last):
File "/home/code/unet_deeplab/RobustVideoMatting/inference.py", line 193, in <module>
converter.convert(
File "/home/code/unet_deeplab/RobustVideoMatting/inference.py", line 169, in convert
convert_video(self.model, device=self.device, dtype=torch.float32, *args, **kwargs)
File "/home/code/unet_deeplab/RobustVideoMatting/inference.py", line 127, in convert_video
fgr, pha, *rec = model(src, *rec, downsample_ratio)
File "/home/anaconda3/envs/bytedance/lib/python3.9/site-packages/torch/nn/modules/module.py", line 1051, in _call_impl
return forward_call(*input, **kwargs)
RuntimeError: The following operation failed in the TorchScript interpreter.
Traceback of TorchScript (most recent call last):
File "/home/code/unet_deeplab/RobustVideoMatting/model/model.py", line 54, in forward
src_sm = src
f1, f2, f3, f4 = self.backbone(src_sm)
~~~~~~~~~~~~~ <--- HERE
f4 = self.aspp(f4)
hid, *rec = self.decoder(src_sm, f1, f2, f3, f4, r1, r2, r3, r4)
File "/home/code/unet_deeplab/RobustVideoMatting/model/resnet.py", line 43, in forward
def forward(self, x):
if x.ndim == 5:
return self.forward_time_series(x)
~~~~~~~~~~~~~~~~~~~~~~~~ <--- HERE
else:
return self.forward_single_frame(x)
File "/home/code/unet_deeplab/RobustVideoMatting/model/resnet.py", line 37, in forward_time_series
def forward_time_series(self, x):
B, T = x.shape[:2]
features = self.forward_single_frame(x.flatten(0, 1))
~~~~~~~~~~~~~~~~~~~~~~~~~ <--- HERE
features = [f.unflatten(0, (B, T)) for f in features]
return features
File "/home/code/unet_deeplab/RobustVideoMatting/model/resnet.py", line 21, in forward_single_frame
def forward_single_frame(self, x):
x = self.conv1(x)
~~~~~~~~~~ <--- HERE
x = self.bn1(x)
x = self.relu(x)
File "/home/anaconda3/envs/bytedance/lib/python3.9/site-packages/torch/nn/modules/conv.py", line 443, in forward
def forward(self, input: Tensor) -> Tensor:
return self._conv_forward(input, self.weight, self.bias)
~~~~~~~~~~~~~~~~~~ <--- HERE
File "/home/anaconda3/envs/bytedance/lib/python3.9/site-packages/torch/nn/modules/conv.py", line 439, in _conv_forward
weight, bias, self.stride,
_pair(0), self.dilation, self.groups)
return F.conv2d(input, weight, bias, self.stride,
~~~~~~~~ <--- HERE
self.padding, self.dilation, self.groups)
RuntimeError: Given groups=1, weight of size [64, 3, 7, 7], expected input[1, 1, 409, 512] to have 3 channels, but got 1 channels instead
Exception ignored in: <function tqdm.__del__ at 0x7f6bcf1e55e0>
Traceback (most recent call last):
File "/home/anaconda3/envs/bytedance/lib/python3.9/site-packages/tqdm/std.py", line 1145, in __del__
File "/home/anaconda3/envs/bytedance/lib/python3.9/site-packages/tqdm/std.py", line 1299, in close
File "/home/anaconda3/envs/bytedance/lib/python3.9/site-packages/tqdm/std.py", line 1492, in display
File "/home/anaconda3/envs/bytedance/lib/python3.9/site-packages/tqdm/std.py", line 1148, in __str__
File "/home/anaconda3/envs/bytedance/lib/python3.9/site-packages/tqdm/std.py", line 1450, in format_dict
TypeError: cannot unpack non-iterable NoneType object我是用的启动命令
python3 inference.py \
--variant resnet50 \
--checkpoint "./checkpoint/rvm_resnet50.pth" \
--device cuda \
--input-source "./input/png_sequence/1次处理/" \
--output-type png_sequence \
--output-composition "./output/png_sequence/test1/composition/" \
--output-alpha "./output/png_sequence/test1/alpha/" \
--output-foreground "./output/png_sequence/test1/foreground/" \
--output-video-mbps 4 \
--seq-chunk 1
调用 convert_video( ),输入为图片路径，输出的图片是透明背景，是否有办法更换背景为黑色背景