metafile.yml 1.0 KB

1234567891011121314151617181920212223242526272829303132333435
  1. Collections:
  2. - Name: SORT
  3. Metadata:
  4. Training Techniques:
  5. - SGD with Momentum
  6. Training Resources: 8x V100 GPUs
  7. Architecture:
  8. - ResNet
  9. - FPN
  10. Paper:
  11. URL: https://arxiv.org/abs/1602.00763
  12. Title: Simple Online and Realtime Tracking
  13. README: configs/sort/README.md
  14. Models:
  15. - Name: sort_faster-rcnn_r50_fpn_8xb2-4e_mot17halftrain_test-mot17halfval
  16. In Collection: SORT
  17. Config: configs/mot/sort/sort_faster-rcnn_r50_fpn_8xb2-4e_mot17halftrain_test-mot17halfval.py
  18. Metadata:
  19. Training Data: MOT17-half-train
  20. inference time (ms/im):
  21. - value: 53.8
  22. hardware: V100
  23. backend: PyTorch
  24. batch size: 1
  25. mode: FP32
  26. resolution: (640, 1088)
  27. Results:
  28. - Task: Multiple Object Tracking
  29. Dataset: MOT17-half-val
  30. Metrics:
  31. MOTA: 62.0
  32. IDF1: 57.8
  33. HOTA: 52.0
  34. Weights: https://download.openmmlab.com/mmtracking/mot/faster_rcnn/faster-rcnn_r50_fpn_4e_mot17-half-64ee2ed4.pth