-
Notifications
You must be signed in to change notification settings - Fork 2k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
posetrack18的复现问题 #1205
Comments
你这个可以正常训练吗? |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
您好,我在尝试复现posetrack18数据集上的跟踪性能时发现结果与论文结果(下图)存在较大差距,不知道是什么问题导致的...
请问是因为pose model #563 没有在posetrack18上finetune的原因吗?还是因为re-id model也需要在posetrack18上再训练?希望可以提供这些finetune后的模型,非常感谢!
我当前采用的tracker_cfg.py文件内容如下:
from easydict import EasyDict as edict
cfg = edict()
cfg.nid = 1000
cfg.arch = "osnet_ain" # "osnet" or "res50-fc512"
cfg.loadmodel = "trackers/weights/osnet_ain_x1_0_msmt17_256x128_amsgrad_ep50_lr0.0015_coslr_b64_fb10_softmax_labsmth_flip_jitter.pth"
cfg.frame_rate = 30
cfg.track_buffer = 240
cfg.conf_thres = 0.5
cfg.nms_thres = 0.4
cfg.iou_thres = 0.5
The text was updated successfully, but these errors were encountered: