hello，I want to test on my own dataset，what parameter should i change，thankyou very much.

skhadem / 3D-BoundingBox

PyTorch implementation for 3D Bounding Box Estimation Using Deep Learning and Geometry

MIT License

435 stars 96 forks source link

hello，I want to test on my own dataset，what parameter should i change，thankyou very much. #6

Open Liqing6668 opened 4 years ago

skhadem commented 4 years ago

How closely does you dataset follow the KITTI format? If it is similar then simply changing the paths would do, but if it is different more work would be required to read in the data.

mheriyanto commented 4 years ago

Hello @skhadem , I try to train my own dataset that followed the KITTI format and I added some classes for classification but i stuck on this file {"cyclist": {"count": 1627, "total": [2826.429999999993, 970.950000000002, 2869.289999999997]}, ...} . What "count": 1627 is number of cyclist on KITTI dataset and "total": [2826.429999999993, 970.950000000002, 2869.289999999997] is real dimension (length, width and, height) of cyclist on meters or feet? Thanks.