Zhongdao Towards-Realtime-MOT issues

Zhongdao / Towards-Realtime-MOT

Joint Detection and Embedding for fast multi-object tracking

MIT License

2.37k stars 539 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

Cityperson Link failure

#258 wwwshengbo opened 2 months ago
0
TypeError: can't convert cuda:0 device type tensor to numpy. Use Tensor.cpu() to copy the tensor to host memory first.

#257 SmallGGgg opened 7 months ago
0
Cityperson.zip cound not be downloaded from google drive

#256 rbler1234 opened 11 months ago
0
mAP 0 on custom dataset for Car tracking

#255 DariusCorp opened 1 year ago
1
Losing Tracking

#254 jaideep11061982 opened 1 year ago
0
Why do I train to get Nan result?

#253 CherryMiya closed 1 year ago
0
ZeroDivisionError: division by zero

#252 jzp-skam opened 1 year ago
3
could I use this model with real-stream like camera?

#251 Wooho-Moon opened 1 year ago
0
can train multi-class Tracking ?

#250 xuaixi opened 2 years ago
1
关于跟踪器update阶段轨迹最后的summarize中self.lost_strack轨迹问题

#249 Xavier-Lin opened 2 years ago
0
Why the embedding representation of appearance is D*H*W, how can so many representations distinguish the representation of positive and negative samples

#248 APeiZou opened 2 years ago
0
1-based indexing MOT dataset

#247 sieumap43 opened 2 years ago
0
What's the correct value of "conf" during training?

#246 JackWoo0831 opened 2 years ago
2
box, conf, id, total, nTargets meaning?

#245 tdchua opened 2 years ago
0
怎么查看损失曲线？How to check loss curve?

#244 MikeMegamind opened 2 years ago
0
图片上没有检测结果

#243 JAYCHOU2020 opened 2 years ago
1
Could find no file with path ?

#242 semihzaman opened 2 years ago
1
Cityperson.zip from google drive error: invalid zip file with overlapped components (possible zip bomb)

#241 yktangac opened 2 years ago
1
输入视频检测时显示图片，未响应

#240 xuxueminglimeng opened 2 years ago
0
The output file error

#239 briansodenkin opened 2 years ago
3
Only use cuhksysu dataset to train, but when i evaluate on mot16 ,there is no detection

#238 Xiangyu33 opened 2 years ago
3
2*RTX3090 , just 6FPS, why?

#237 zhaotun opened 2 years ago
1
For the BDD100K dataset (multi-classes multi-object tracking), is there support for this project?

#236 long2double opened 2 years ago
1
MOT17 zip corrupted

#235 shreejalt opened 2 years ago
1
Why deal with new (unconfirmed) tracklets by only using box IoU?

#234 Johnqczhang opened 2 years ago
0
A minor bug

#233 Johnqczhang opened 2 years ago
0
matching策略中，对re-id应该默认采用cosine相似度，为什么代码里写的是欧式距离

#232 Bilibilee opened 2 years ago
1
Ship tracking

#231 fancyfeifei opened 2 years ago
2
Division by zero error when running demo

#230 amitgalor18 opened 2 years ago
4
how to evaluate ?

#229 ssbilakeri opened 2 years ago
0
why i got nan result?

#228 datuizhuang closed 3 years ago
1
training and inference preprocessing (255.0 scaling)

#227 ZiyanZhu1994 opened 3 years ago
0
Video Resolution 1280X720, 没有对应的训练好的model, 性能会影响很大么？多谢

#226 xukaiquan opened 3 years ago
0
训练时需要进行shuffle操作吗？

#225 xdg2016 opened 3 years ago
0
Got error when using yolov3-tiny.conv.15 and yolov3-tiny.cfg

#224 mheriyanto closed 3 years ago
4
AttributeError: 'LoadImages' object has no attribute 'files'

#223 fanshu4869 opened 3 years ago
4
none output

#222 geyu1998 opened 3 years ago
0
Is there a model zoo with pre-trained models?

#221 athulvingt opened 3 years ago
1
C++ implementation of JDETracker

#220 yongjingli opened 3 years ago
0
Storage has wrong size: expected 0 got 1024

#219 nhanbka opened 3 years ago
0
Problem with re-ID features and fuse_motion

#218 gouttham closed 3 years ago
2
Dose anyone have a try to reproduce the model by a 60-epoch training process？

#217 scuShaker opened 3 years ago
1
Training loss suddenly became nan on my custom dataset. What is the problem?

#216 AlekseySh opened 3 years ago
2
The MOT16 train set results in README.md can not be reproduce?

#215 nemonameless opened 3 years ago
3
can not get 74.8 MOTA in MOT-16 trainging set

#214 scuShaker opened 3 years ago
6
sample integration

#213 partus opened 3 years ago
0
id in train loss is too high. Is this normal?

#212 zhangyihao91 opened 3 years ago
2
training loss is nan

#211 wulongjian opened 3 years ago
3
在RTX2080TI上面，FPS是8FPS，但是作者给的是22.2FPS，请问为啥不一样。

#210 nimochuanzhang closed 3 years ago
4
Poor results on rodent videos

#209 catubc opened 3 years ago
0