IndexError: too many indices for tensor of dimension 1

使用YoloV5和DeepSort做目标检测跟踪的时候，将视频流切换到了树莓派的网络摄像头，报以下错误

Traceback (most recent call last):
  File ".demo_simple.py", line 77, in <module>
    demo.detect()
  File ".demo_simple.py", line 58, in detect
    track_bboxes = tracker_l.update_tracker(frame, yolo_bboxes)
  File "I:DocumentsMachine Learning基础趣味CV项目实战项目7：YOLOv5+Deepsort 道路车辆分析2.deepsorthighway_detection.py", line 133, in update_tracker
    outputs = self.deepsort.update(xywhs, confss, clss, image)
  File "I:DocumentsMachine Learning基础趣味CV项目实战项目7：YOLOv5+Deepsort 道路车辆分析2.deepsortdeep_sortdeep_sortdeep_sort.py", line 32, in update
    bbox_tlwh = self._xywh_to_tlwh(bbox_xywh)
  File "I:DocumentsMachine Learning基础趣味CV项目实战项目7：YOLOv5+Deepsort 道路车辆分析2.deepsortdeep_sortdeep_sortdeep_sort.py", line 55, in _xywh_to_tlwh
    bbox_tlwh[:, 0] = bbox_xywh[:, 0] - bbox_xywh[:, 2]/2.
IndexError: too many indices for tensor of dimension 1

最后在GitHub找到了原因

原因是当前画面中检测框数目为空