Deep-Flow-Guided-Video-Inpainting
Deep-Flow-Guided-Video-Inpainting copied to clipboard
nbei
SIGSEGV: Segmentation fault
A “SIGSEGV: Segmentation fault” error occurred while running video_inpaint.py
Hi! I locate the error at tools/infer_flownet2.py at 53 line 'flow = Flownet(f1, f2)', I want to know why are you passing in two parameters for the Flownet? What does the Flownet stand for?Does it not represent the optical flow produced by FlowNet2? Thank you very much.
(deep_flow) $ python tools/video_inpaint.py --frame_dir ./demo/frames --MASK_ROOT ./demo/masks --img_size 512 832 --FlowNet2 --DFC --ResNet101 --Propagation before extract_flow extract_flow... This Video has 80 Images The optical flow list has been generated: ./demo/video_flow.txt before loading pretrained_model_flownet2 ====> Loading ./pretrained_models/FlowNet2_checkpoint.pth.tar after loading pretrained_model_flownet2 type = <class 'torch.utils.data.dataloader.DataLoader'> 158 before 第一个Good! Good! <class 'torch.Tensor'> torch.Size([1, 3, 512, 832]) before Flownet /home/yf/.local/share/virtualenvs/deep_flow-NGK_eXwf/lib/python3.6/site-packages/torch/nn/functional.py:1749: UserWarning: Default upsampling behavior when mode=bilinear is changed to align_corners=False since 0.4.0. Please specify align_corners=True if the old behavior is desired. See the documentation of nn.Upsample for details. "See the documentation of nn.Upsample for details.".format(mode)) Segmentation fault
Hi! I locate the error at tools/infer_flownet2.py at 53 line 'flow = Flownet(f1, f2)', I want to know why are you passing in two parameters for the Flownet? What does the Flownet stand for?Does it not represent the optical flow produced by FlowNet2? Thank you very much.
(deep_flow) $ python tools/video_inpaint.py --frame_dir ./demo/frames --MASK_ROOT ./demo/masks --img_size 512 832 --FlowNet2 --DFC --ResNet101 --Propagation before extract_flow extract_flow... This Video has 80 Images The optical flow list has been generated: ./demo/video_flow.txt before loading pretrained_model_flownet2 ====> Loading ./pretrained_models/FlowNet2_checkpoint.pth.tar after loading pretrained_model_flownet2 type = <class 'torch.utils.data.dataloader.DataLoader'> 158 before 第一个Good! Good! <class 'torch.Tensor'> torch.Size([1, 3, 512, 832]) before Flownet /home/yf/.local/share/virtualenvs/deep_flow-NGK_eXwf/lib/python3.6/site-packages/torch/nn/functional.py:1749: UserWarning: Default upsampling behavior when mode=bilinear is changed to align_corners=False since 0.4.0. Please specify align_corners=True if the old behavior is desired. See the documentation of nn.Upsample for details. "See the documentation of nn.Upsample for details.".format(mode)) Segmentation fault
hi, @puhan123 i found this is caused by cv2 requires libSM.so.6, you can try 'apt-get install -y python-qt4'
