computervisioneng / object-tracking-yolov8-deep-sort

object-tracking-yolov8-deep-sort

Yolov8 object detection + deep sort object tracking !

requirements

• Python 3.7
• requirements.txt

Deep Sort

We are working on this fork from deep sort official implementation.

You can download deep sort feature extraction model here.

data

You can download the same data I use in the video here.

object-tracking-yolov8-deep-sort's Issues

In the tracker class you created, you declared the cosine distance used for distance measurement, but in deepsort, cosine distance and Mahalanobis distance were combined. I'm not sure if this confusion arose because I didn't fully understand the code, or if this issue does exist

graph_def.ParseFromString(file_handle.read()),error UnicodeDecodeError: 'utf-8' codec can't decode byte 0xd5 in position 77: invalid continuation byte

video path and out path .\data\people.mp4 .\out.mp4
2023-03-06 19:00:22.905833: I tensorflow/core/platform/cpu_feature_guard.cc:193] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations: AVX AVX2
To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.
Traceback (most recent call last):
File "F:\object-tracking-yolov8-deep-sort-master\main.py", line 22, in
tracker = Tracker()
File "F:\object-tracking-yolov8-deep-sort-master\tracker.py", line 21, in init
self.encoder = gdet.create_box_encoder(encoder_model_filename, batch_size=1)
File "F:\object-tracking-yolov8-deep-sort-master\deep_sort\tools\generate_detections.py", line 100, in create_box_encoder
image_encoder = ImageEncoder(model_filename, input_name, output_name)
File "F:\object-tracking-yolov8-deep-sort-master\deep_sort\tools\generate_detections.py", line 78, in init
graph_def.ParseFromString(file_handle.read())
File "D:\Users\liush\Anaconda3\envs\yolov8\lib\site-packages\tensorflow\python\lib\io\file_io.py", line 116, in read
self._preread_check()
File "D:\Users\liush\Anaconda3\envs\yolov8\lib\site-packages\tensorflow\python\lib\io\file_io.py", line 77, in _preread_check
self._read_buf = _pywrap_file_io.BufferedInputStream(
UnicodeDecodeError: 'utf-8' codec can't decode byte 0xd5 in position 77: invalid continuation byte

ModuleNotFoundError: No module named 'deep_sort'

I have cloned the repository of deep_sort but it always shows the message 'ModuleNotFoundError: No module named 'deep_sort''

scikit-learn == 0.21.0 not found

ERROR: Could not find a version that satisfies the requirement scikit-learn==0.21.0 (from versions: 0.9, 0.10, 0.11, 0.12, 0.12.1, 0.13, 0.13.1, 0.14, 0.14.1, 0.15.0, 0.15.1, 0.15.2, 0.16.0, 0.16.1, 0.17, 0.17.1, 0.18, 0.18.1, 0.18.2, 0.19.0, 0.19.1, 0.19.2, 0.20.0, 0.20.1, 0.20.2, 0.20.3, 0.20.4, 0.21.1, 0.21.2, 0.21.3, 0.22, 0.22.1, 0.22.2.post1, 0.23.0, 0.23.1, 0.23.2, 0.24.0, 0.24.1, 0.24.2, 1.0, 1.0.1, 1.0.2, 1.1.0, 1.1.1, 1.1.2, 1.1.3, 1.2.0rc1, 1.2.0, 1.2.1)
ERROR: No matching distribution found for scikit-learn==0.21.0

python == 3.10

Tracker.update(frame, detections) generates error when zero length detections is passed (no detection on the current frame)

File "/Users/[path]/tracker.py", line 30, in update
bboxes[:, 2:] = bboxes[:, 2:] - bboxes[:, 0:2]
IndexError: too many indices for array: array is 1-dimensional, but 2 were indexed

It happens whenever there is no detection, guess need to add a check for empty detections list.

scikit-learn version in requirements.txt is wrong

in requirement file, we have scikit-learn==0.21.0.

running pip install -r requirements.txt there's an error:

ERROR: Could not find a version that satisfies the requirement scikit-learn==0.21.0 (from versions: 0.9, 0.10, 0.11, 0.12, 0.12.1, 0.13, 0.13.1, 0.14, 0.14.1, 0.15.0, 0.15.1, 0.15.2, 0.16.0, 0.16.1, 0.17, 0.17.1, 0.18, 0.18.1, 0.18.2, 0.19.0, 0.19.1, 0.19.2, 0.20.0, 0.20.1, 0.20.2, 0.20.3, 0.20.4, 0.21.1, 0.21.2, 0.21.3, 0.22, 0.22.1, 0.22.2.post1, 0.23.0, 0.23.1, 0.23.2, 0.24.0, 0.24.1, 0.24.2, 1.0, 1.0.1, 1.0.2, 1.1.0, 1.1.1, 1.1.2, 1.1.3, 1.2.0rc1, 1.2.0, 1.2.1, 1.2.2)
ERROR: No matching distribution found for scikit-learn==0.21.0

Error ModuleNotFoundError: No module named 'sklearn.utils.linear_assignment_'

Traceback (most recent call last):
File "c:\Programming\opencv\yolo\dbscan_deepsort.py", line 9, in
from tracker import Tracker
File "c:\Programming\opencv\yolo\tracker.py", line 1, in
from deep_sort.deep_sort.tracker import Tracker as DeepSortTracker
File "c:\Programming\opencv\yolo\deep_sort\deep_sort\tracker.py", line 5, in
from . import linear_assignment
File "c:\Programming\opencv\yolo\deep_sort\deep_sort\linear_assignment.py", line 4, in
from sklearn.utils.linear_assignment_ import linear_assignment
ModuleNotFoundError: No module named 'sklearn.utils.linear_assignment_'

From what I find online linear_assignment is discontinued in newer scikit-learn versions and I'm unable to install the version in requirments file.
ERROR: Could not find a version that satisfies the requirement scikit-learn==0.21.0 (from versions: 0.9, 0.10, 0.11, 0.12, 0.12.1, 0.13, 0.13.1, 0.14, 0.14.1, 0.15.0, 0.15.1, 0.15.2, 0.16.0, 0.16.1, 0.17, 0.17.1, 0.18, 0.18.1, 0.18.2, 0.19.0, 0.19.1, 0.19.2, 0.20.0, 0.20.1, 0.20.2, 0.20.3, 0.20.4, 0.21.1, 0.21.2, 0.21.3, 0.22, 0.22.1, 0.22.2.post1, 0.23.0, 0.23.1, 0.23.2, 0.24.0, 0.24.1, 0.24.2, 1.0, 1.0.1, 1.0.2, 1.1.0, 1.1.1, 1.1.2, 1.1.3, 1.2.0rc1, 1.2.0, 1.2.1, 1.2.2, 1.3.0rc1)
ERROR: No matching distribution found for scikit-learn==0.21.0

If I install for some even older version I have error with sklearn.cluster.dbscan which I intend to use in the same program as deepsort. Is there any way to make deepsort work with different scikit-learn version?

Fall detection using Yolov8-pose model

Do you have any project of fall detection with Yolov8-pose model. Please provide the details if you have.

How to deal with the question : cannot connect to X server

(openmmlab) guanhaoliang@scut-X99P-SLI:~/mmpose$ python demo/topdown_demo_with_mmdet.py demo/mmdetection_cfg/rtmdet_m_640-8xb32_coco-person.py https://download.openmmlab.com/mmpose/v1/projects/rtmpose/rtmdet_m_8xb32-100e_coco-obj365-person-235e8209.pth configs/body_2d_keypoint/rtmpose/body8/rtmpose-m_8xb256-420e_body8-256x192.py https://download.openmmlab.com/mmpose/v1/projects/rtmposev1/rtmpose-m_simcc-body7_pt-body7_420e-256x192-e48f03d0_20230504.pth --input tests/data/coco/000000197388.jpg --show --draw-heatmap --output-root vis_results/
Loads checkpoint by http backend from path: https://download.openmmlab.com/mmpose/v1/projects/rtmpose/rtmdet_m_8xb32-100e_coco-obj365-person-235e8209.pth
Loads checkpoint by http backend from path: https://download.openmmlab.com/mmpose/v1/projects/rtmposev1/rtmpose-m_simcc-body7_pt-body7_420e-256x192-e48f03d0_20230504.pth
/home/guanhaoliang/.conda/envs/openmmlab/lib/python3.8/site-packages/torch/functional.py:504: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at /opt/conda/conda-bld/pytorch_1699449185072/work/aten/src/ATen/native/TensorShape.cpp:3526.)
return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined]
/home/guanhaoliang/mmpose/mmpose/models/heads/coord_cls_heads/rtmcc_head.py:217: UserWarning: The predicted simcc values are normalized for visualization. This may cause discrepancy between the keypoint scores and the 1D heatmaps.
warnings.warn('The predicted simcc values are normalized for '
: cannot connect to X server

Which python Version should I use?

In the requirements, you told to install python 3.7. But it will no longer be supported by the Python Software Foundation (PSF) starting June 27, 2023. Which version should I use now?
If I use later version of python, will be any problems to run the code?