系列文章目录
前言
一、环境配置
碰到的主要问题是包的版本匹配问题。
具体安装可见如下链接,但是在安装时还是遇到包版本不匹配问题,在本人电脑环境下,包安装与版本如下:
OpenMMLab MMTracking目标跟踪环境搭建(一)
pip install mmengine -f https://download.pytorch.org/whl/lts/1.8/torch_lts.html -i http://mirrors.aliyun.com/pypi/simple/ --trusted-host mirrors.aliyun.co
pip install mmcv-full==1.6.1 -f https://download.openmmlab.com/mmcv/dist/cu111/torch1.8.0/index.html
pip install asynctest==0.13.0
pip install attributee==0.1.8
pip install attrs==23.1.0
pip install bidict==0.22.1
pip install cachetools==5.3.2
pip install certifi==2023.11.17
pip install charset-normalizer==3.3.2
pip install click==8.1.7
pip install codecov==2.1.13
pip install colorama==0.4.6
pip install coverage==7.3.2
pip install Cython==3.0.5
pip install dominate==2.9.0
pip install dotty-dict==1.3.1
pip install einops==0.7.0
pip install exceptiongroup==1.2.0
pip install flake8==6.1.0
pip install idna==3.4
pip install importlib-metadata==6.8.0
pip install importlib-resources==6.1.1
pip install iniconfig==2.0.0
pip install interrogate==1.5.0
pip install isort==4.3.21
pip install joblib==1.3.2
pip install jsonschema==4.20.0
pip install jsonschema-specifications==2023.11.1
pip install kiwisolver==1.4.5
pip install kwarray==0.6.16
pip install lap==0.4.0
pip install llvmlite==0.41.1
pip install Markdown==3.5.1
pip install mccabe==0.7.0
pip install mdurl==0.1.2
pip install mmcls==0.25.0
pip install mmdet==2.28.2
pip install model-index==0.1.11
pip install modelindex==0.0.2
pip install motmetrics==1.4.0
pip install numba==0.58.1
pip install opencv-python-headless==4.1.2.30
pip install ordered-set==4.1.0
pip install pandas==1.3.5
pip install phx-class-registry==4.0.6
pip install pkgutil_resolve_name==1.3.10
pip install pluggy==1.3.0
pip install py==1.11.0
pip install pycocotools==2.0.7
pip install pycodestyle==2.11.1
pip install pyflakes==3.1.0
pip install PyLaTeX==1.4.2
pip install pytest==7.4.3
pip install pytz==2023.3.post1
pip install referencing==0.31.0
pip install requests==2.31.0
pip install rpds-py==0.13.1
pip install scikit-learn==1.3.2
pip install scipy==1.7.3
pip install seaborn==0.13.0
pip install tabulate==0.9.0
pip install terminaltables==3.1.10
pip install threadpoolctl==3.2.0
pip install toml==0.10.2
pip install tqdm==4.66.1
pip install typing_extensions==4.5.0
pip install tzdata==2023.3
pip install ubelt==1.3.4
pip install urllib3==2.1.0
pip install xdoctest==0.10.0
pip install xmltodict==0.13.0pip install yapf==0.40.0
pip install git+https://github.com/JonathonLuiten/TrackEval.git
pip install git+https://github.com/votchallenge/toolkit.git
pip install git+https://github.com/lvis-dataset/lvis-api.git
pip install git+https://github.com/TAO-Dataset/tao.git
Package Version
------------------------- ------------
addict 2.4.0
asynctest 0.13.0
attributee 0.1.8
attrs 23.1.0
bidict 0.22.1
cachetools 5.3.2
certifi 2023.11.17
charset-normalizer 3.3.2
click 8.1.7
codecov 2.1.13
colorama 0.4.6
contourpy 1.1.1
coverage 7.3.2
cycler 0.12.1
Cython 3.0.5
dominate 2.9.0
dotty-dict 1.3.1
einops 0.7.0
exceptiongroup 1.2.0
flake8 6.1.0
fonttools 4.45.0
idna 3.4
importlib-metadata 6.8.0
importlib-resources 6.1.1
iniconfig 2.0.0
interrogate 1.5.0
isort 4.3.21
joblib 1.3.2
jsonschema 4.20.0
jsonschema-specifications 2023.11.1
kiwisolver 1.4.5
kwarray 0.6.16
lap 0.4.0
llvmlite 0.41.1
Markdown 3.5.1
markdown-it-py 3.0.0
matplotlib 3.7.4
mccabe 0.7.0
mdurl 0.1.2
mmcls 0.25.0
mmdet 2.28.2
mmengine 0.10.1
model-index 0.1.11
modelindex 0.0.2
motmetrics 1.4.0
numba 0.58.1
numpy 1.22.4
opencv-python 4.8.1.78
opencv-python-headless 4.1.2.30
ordered-set 4.1.0
packaging 23.2
pandas 1.3.5
phx-class-registry 4.0.6
Pillow 9.5.0
pip 23.0.1
pkgutil_resolve_name 1.3.10
platformdirs 4.0.0
pluggy 1.3.0
py 1.11.0
pycocotools 2.0.7
pycodestyle 2.11.1
pyflakes 3.1.0
Pygments 2.17.2
PyLaTeX 1.4.2
pyparsing 3.1.1
pytest 7.4.3
python-dateutil 2.8.2
pytz 2023.3.post1
PyYAML 6.0.1
referencing 0.31.0
requests 2.31.0
rich 13.7.0
rpds-py 0.13.1
scikit-learn 1.3.2
scipy 1.7.3
seaborn 0.13.0
setuptools 66.0.0
six 1.16.0
tabulate 0.9.0
termcolor 2.3.0
terminaltables 3.1.10
threadpoolctl 3.2.0
toml 0.10.2
tomli 2.0.1
torch 1.8.1+cu111
torchaudio 0.8.1
torchvision 0.9.1+cu111
tqdm 4.66.1
typing_extensions 4.5.0
tzdata 2023.3
ubelt 1.3.4
urllib3 2.1.0
wheel 0.38.4
xdoctest 0.10.0
xmltodict 0.13.0
yapf 0.40.2
zipp 3.17.0
二、demo_mot_vis.py 运行
需要配参数:
--input
../demo/demo.mp4
--output
mot.mp4
--config
../configs/mot/deepsort/sort_faster-rcnn_fpn_4e_mot17-private.py
三、问题
3.1 运行test.py
配置:
../configs/sot/siamese_rpn/siamese_rpn_r50_20e_vot2018.py
--checkpoint
../checkpoints/epoch_1.pth
--out
results.pkl
--eval
track
问题:
if is_special(region, Special.UNKNOWN) and ignore_unknown:
AttributeError: type object 'Special' has no attribute 'UNKNOWN'
原因: 参数配置出错(注意使用quick_run.md)
- `EVAL_METRICS`:用于评估结果的指标。允许的值取决于数据集,例如, `bbox` 适用于 ImageNet VID, `track` 适用于 LaSOT, `bbox` and `track` 都适用于 MOT17。MMTracking的配置,给出了环境配置的一个参考。
正确
../configs/sot/siamese_rpn/siamese_rpn_r50_20e_vot2018.py
--checkpoint
../checkpoints/epoch_1.pth
--out
results.pkl