Comments (19)
from yolo_tracking.
Should be fixed. Let me know π
from yolo_tracking.
Hi @mikel-brostrom.
I have faced the same issue. After cloning your newer Release v10.0.60 another problem has emerged, as follows:
File "/home/arash/TAD/ObjectTrack/Yolov8_mikel-brostrom/yolo_tracking/boxmot/motion/kalman_filters/deepocsort_kf.py", line 506, in update
self.history_obs.append(z)
AttributeError: 'NoneType' object has no attribute 'append'
from yolo_tracking.
Hi @mikel-brostrom. I have faced the same issue. After cloning your newer Release v10.0.60 another problem has emerged, as follows:
File "/home/arash/TAD/ObjectTrack/Yolov8_mikel-brostrom/yolo_tracking/boxmot/motion/kalman_filters/deepocsort_kf.py", line 506, in update self.history_obs.append(z) AttributeError: 'NoneType' object has no attribute 'append'
Me too.
from yolo_tracking.
Oh, I missed to replace the last indexing in DeepOCSORT KF
from yolo_tracking.
Try again please π
from yolo_tracking.
Try again please π
Great! The problem is solved. Thank you very much!
from yolo_tracking.
This:
File "/data/users/johannes.geibel/miniconda3/envs/yolo/lib/python3.11/site-packages/boxmot/motion/kalman_filters/ocsort_kf.py", line 400, in unfreeze
index1 = indices[-2]
~~~~~~~^^^^
IndexError: index -2 is out of bounds for axis 0 with size 1
Should be now fixed as well @johannesgeibel
from yolo_tracking.
Try again please π
Thanks mikel. It works fine, but I recommend those who face similar issue to remove the environment and redefine all packages and dependencies again in new environment. This will avoid confusions.
from yolo_tracking.
This:
File "/data/users/johannes.geibel/miniconda3/envs/yolo/lib/python3.11/site-packages/boxmot/motion/kalman_filters/ocsort_kf.py", line 400, in unfreeze
index1 = indices[-2]IndexError: index -2 is out of bounds for axis 0 with size 1
Should be now fixed as well @johannesgeibel
@mikel-brostrom Sadly not...
I printed new_history
, occur
, and indices
when the error occurs:
[None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, array([[7.94601074e+02],
[8.38841797e+02],
[2.67891208e+04],
[6.00677161e-01]])]
[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0]
[49]
The reason is that there is only observation 50 not None and thus len(indices) == 1
, causing the error. Could this be due to the dques being fixed at length 50, and I do not have observations over a longer time period (while max_age
being set to 50 as well? When I reduced max_age
to 30, the error did not occur at this frame anymore.
I am not sure where the need for the maximum length of 50 for the history observations comes from, besides probably memory efficiency. But wouldn't it be a possibility, to set the length to max_age + x
, ensuring that it is always long enougth?
from yolo_tracking.
Your are right. The max_age
parameter cannot be longer than the history
from yolo_tracking.
Could this be due to the dques being fixed at length 50, and I do not have observations over a longer time period (while max_age being set to 50 as well?
Aha, that is why I couldn't find this bug π€£. In the repo max_age
is set to 30. I think an assertion here would be enough
from yolo_tracking.
The same bug
File "/media/chenyi/Elements/Tools/yolo_tracking-master/boxmot/motion/kalman_filters/hybridsort_kf.py", line 399, in unfreeze index1 = indices[-2] IndexError: index -2 is out of bounds for axis 0 with size 1
is happening to HybridSORT, though the line seems to have been updated according to previous comments self.history_obs = deque(list(self.history_obs)[:-1])
.
Any idea on how to solve this bug?
from yolo_tracking.
This was updated in the last release @ch3ny1. Check out master
from yolo_tracking.
This was updated in the last release @ch3ny1. Check out master
The bug is still there,when i run the last release code!
ile "/mnt/e/python_project_new/yolo_tracking-10.0.65/boxmot/utils/init.py", line 79, in wrapper
tracks = self.update(instance, dets, im)
File "/mnt/e/python_project_new/yolo_tracking-10.0.65/boxmot/trackers/hybridsort/hybridsort.py", line 469, in update
self.active_tracks[m[1]].update(dets[m[0], :], dets0[m[0], 5], dets0[m[0], 6], id_feature_keep[m[0], :])
File "/mnt/e/python_project_new/yolo_tracking-10.0.65/boxmot/trackers/hybridsort/hybridsort.py", line 286, in update
self.kf.update(convert_bbox_to_z(bbox))
File "/mnt/e/python_project_new/yolo_tracking-10.0.65/boxmot/motion/kalman_filters/hybridsort_kf.py", line 487, in update
self.unfreeze()
File "/mnt/e/python_project_new/yolo_tracking-10.0.65/boxmot/motion/kalman_filters/hybridsort_kf.py", line 399, in unfreeze
index1 = indices[-2]
IndexError: index -2 is out of bounds for axis 0 with size 1
from yolo_tracking.
What have you set max_age
to? @TSLNIHAOGIT
from yolo_tracking.
I have tried max_age == 10, 20 and 30 and the same traceback
from yolo_tracking.
The same bug
File "/media/chenyi/Elements/Tools/yolo_tracking-master/boxmot/motion/kalman_filters/hybridsort_kf.py", line 399, in unfreeze index1 = indices[-2] IndexError: index -2 is out of bounds for axis 0 with size 1
is happening to HybridSORT, though the line seems to have been updated according to previous commentsself.history_obs = deque(list(self.history_obs)[:-1])
.Any idea on how to solve this bug?
same problem
from yolo_tracking.
I faced the same problem. Is there a solution?
from yolo_tracking.
Related Issues (20)
- StrongSORT plotting results HOT 4
- TensorRT HOT 9
- Applicability of Boxmot in my case HOT 6
- `KalmanBoxTracker.observations` grows infinitely, causing out of memory HOT 2
- Unable to pass `embs` to tarcker HOT 3
- it can't run when i start tracking
- How can i use ZED camera for this approach? HOT 3
- (OCSORT, DeepOCSORT) shape specs not match doc function HOT 4
- Why reid is not working at all in my case? HOT 5
- Disable ReID module HOT 3
- Using different Reid models to evaluate DeepOcsort, but getting the same evaluation results HOT 4
- PyTorch version automatic update HOT 2
- Evaluation using bytetrack HOT 2
- Why does the frame id of the txt file generated by the evaluation start at 2 HOT 17
- calculate FPS HOT 2
- How to divide the MOT17 or MOT20 dataset? HOT 1
- more reid HOT 3
- (yolov8track) PS F:\yolov8_tracking> python tracking/evolve.py --benchmark MOT17-mini --dets runs/dets_n_embs/yolov8n/dets/MVI_20011.txt --embs runs/dets_n_embs/yolov8n/embs/resnet50_market1501 --n-trials 9 --tracking-method botsort --source ./asse ts/MOT17-mini/train C:\Users\79223\AppData\Local\anaconda3\envs\yolov8track\lib\site-packages\_distutils_hack\__init__.py:33: UserWarning: Setuptools is replacing distutils. warnings.warn("Setuptools is replacing distutils.") Traceback (most recent call last): File "F:\yolov8_tracking\tracking\evolve.py", line 138, in <module> opt.yolo_model = [Path(y).resolve() for y in opt.yolo_model] TypeError: 'WindowsPath' object is not iterabl
- ε¨δΈδΈͺιε₯ζ₯εε°θ―ζδ½ HOT 1
- Full MOT17 evaluation results on all trackers
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
π Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. πππ
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google β€οΈ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from yolo_tracking.