In the code currently the class_name_valid = class_names_valid = ["pedestrian", "rider", "car", "truck", "bus", "train", "motorcycle", "bicycle"] which is less than the number of classes the BDD MOT dataset provides.
Please let me know if there is any other information needed in order to resolve this. I just followed the steps provided in the repository.
Thanks in advance.
root@eb894c76f144:/workspace/cusfair/src# CUDA_VISIBLE_DEVICES=0 python -W ignore track_bdd.py mot --val_bdd100k True --load_model ../exp/mot/bdd_hrnet/model_25.pth --conf_thres 0.4 --arch 'hrnet_32' --reid_dim 128 --exp_id bdd_hrnet --K 256 Fix size testing. training chunk_sizes: [6, 6] The output will be saved to /workspace/cusfair/src/lib/../../exp/mot/bdd_hrnet heads {'hm': 11, 'wh': 2, 'id': 128, 'reg': 2} Namespace(K=256, arch='hrnet_32', batch_size=12, cat_spec_wh=False, chunk_sizes=[6, 6], conf_thres=0.4, data_cfg='../src/lib/cfg/bdd100k.json', data_dir='../data/MOT', dataset='jde', debug_dir='/workspace/cusfair/src/lib/../../exp/mot/bdd_hrnet/debug', dense_wh=False, det_thres=0.3, down_ratio=4, exp_dir='/workspace/cusfair/src/lib/../../exp/mot', exp_id='bdd_hrnet', fix_res=True, gpus=[0, 1], gpus_str='0, 1', head_conv=256, heads={'hm': 11, 'wh': 2, 'id': 128, 'reg': 2}, hide_data_time=False, hm_weight=1, id_loss='ce', id_weight=1, img_size=(1088, 608), input_h=608, input_res=1088, input_video='../videos/MOT16-03.mp4', input_w=1088, keep_res=False, load_model='../exp/mot/bdd_hrnet/model_25.pth', lr=0.0001, lr_step=[20, 27], master_batch_size=6, mean=None, metric='loss', min_box_area=200, mse_loss=False, nID=112820, nms_thres=0.4, norm_wh=False, not_cuda_benchmark=False, not_prefetch_test=False, not_reg_offset=False, num_classes=11, num_epochs=30, num_iters=-1, num_stacks=1, num_workers=8, off_weight=1, output_format='video', output_h=152, output_res=272, output_root='../results', output_w=272, pad=31, print_iter=0, reg_loss='l1', reg_offset=True, reid_dim=128, resume=False, root_dir='/workspace/cusfair/src/lib/../..', save_all=False, save_dir='/workspace/cusfair/src/lib/../../exp/mot/bdd_hrnet', seed=317, std=None, task='mot', test=False, test_mot15=False, test_mot16=False, test_mot17=False, test_mot20=False, track_buffer=30, trainval=False, val_bdd100k='True', val_intervals=5, val_mot15=False, val_mot16=False, val_mot17=False, val_mot20=False, vis_thresh=0.5, wh_weight=0.1) 2021-07-14 09:12:31 [INFO]: start seq: bdd Creating model... loaded ../exp/mot/bdd_hrnet/model_25.pth, epoch 25 2021-07-14 09:12:41 [INFO]: Processing frame 0 (100000.00 fps) Creating model... loaded ../exp/mot/bdd_hrnet/model_25.pth, epoch 25 Traceback (most recent call last): File "track_bdd.py", line 248, in
save_videos=False)
File "track_bdd.py", line 197, in main
save_dir=output_dir, show_image=show_image, frame_rate=frame_rate)
File "track_bdd.py", line 130, in eval_seq
online_targets = tracker.update_sep3(blob, img0, conf_thres=[0.4 ,0.3, 0.4, 0.4, 0.4, 0.4, 0.4, 0.5, 0.5]) ## use class-separated tracker
File "/workspace/cusfair/src/lib/tracker/multitracker.py", line 544, in update_sep3
matches, u_track, u_detection = matching.linear_assignment(dists, thresh=emb_thresh[i_class-1])
IndexError: list index out of range
In the code currently the class_name_valid = class_names_valid = ["pedestrian", "rider", "car", "truck", "bus", "train", "motorcycle", "bicycle"] which is less than the number of classes the BDD MOT dataset provides.
Please let me know if there is any other information needed in order to resolve this. I just followed the steps provided in the repository. Thanks in advance.