How to track visualization on BDD

[NanH5837]

I tried python tools/test.py ${CONFIG_FILE} ${CHECKPOINT_FILE} [--out ${RESULT_FILE}] [--eval ${EVAL_METRICS}] [--show] [--show-dir ${SHOW_DIR}] [--cfg-options] but I didn't get the results of track visualization on BDD

[XiaLiPKU]

Currently, you can resort to this process:

1. Convert the prediction into BDD100K format: https://github.com/SysCV/qdtrack/blob/master/docs/GET_STARTED.md#conversion-to-the-scalabelbdd100k-format
2. Visualize the prediction through bdd100k tools: https://doc.bdd100k.com/usage.html#understanding-the-data

[taheranjary]

Hi there. I ran your pre-trained BDD100K model on the tracking set and saved the results to result.pkl. Then I ran the command python tools/to_bdd100k.py ./configs/bdd100k/qdtrack-frcnn_r50_fpn_12e_bdd100k.py --res results.pkl --task box_track --bdd-dir conversions --nproc 1 --coco-file COCO_PRED_FILE. This script complains saying the file should be a list. If I convert it to a list, consequent scripts complain that the object has no .items(). So what should the result file be, a dictionary or a list?

The errors arise from: qdtrack/tools/to_bdd100k.py line 66 dataset.format_results(results, jsonfile_prefix=args.coco_file) AND qdtrack/qdtrack/core/to_bdd100k/transforms.py, box_track_to_bdd100k() ,for id_, instance in track_dict.items()

[OceanPang]

New inference code available at here.