search

mkocabas / VIBE

Official implementation of CVPR2020 paper "VIBE: Video Inference for Human Body Pose and Shape Estimation"
https://arxiv.org/abs/1912.05656
Other
2.86k stars 550 forks source link

[BUG] Load Darknet Weights Error #246

Open ChawDoe opened 2 years ago

ChawDoe commented 2 years ago 
  File "demo.py", line 414, in <module>
    main(args)
  File "demo.py", line 95, in main
    yolo_img_size=args.yolo_img_size,
  File "/mnt/zhoudeyu/project/save_video/dengyuanzhang/VIBE-master/multi-person-tracker-master/multi_person_tracker/mpt.py", line 53, in __init__
    device=self.device, img_size=yolo_img_size, person_detector=True, video=True, return_dict=True
  File "/mnt/zhoudeyu/project/save_video/dengyuanzhang/VIBE-master/yolov3-pytorch-master/yolov3/yolo.py", line 51, in __init__
    self.model.load_darknet_weights(weights_path)
  File "/mnt/zhoudeyu/project/save_video/dengyuanzhang/VIBE-master/yolov3-pytorch-master/yolov3/models.py", line 315, in load_darknet_weights
    conv_w = torch.from_numpy(weights[ptr : ptr + num_w]).view_as(conv_layer.weight)
RuntimeError: shape '[128, 64, 3, 3]' is invalid for input of size 7246

Here is the information of the bug.