0ssamaak0 / dlta-ai Goto Github PK

• it creates one with the chosen label, and one with the label sam instance
  

having another models (YOLOv8 and SAM) ➡️ we must store checkpoints in a neutral directory

Error message

  File "...\labelmm\labelme-master\labelme\app.py", line 2169, in load_objects_from_json
    for i in range(self.TOTAL_VIDEO_FRAMES)]
AttributeError: 'MainWindow' object has no attribute 'TOTAL_VIDEO_FRAMES'

Error message

line 3194, in deleteSelectedShape
    from_frame.setRange(1, self.TOTAL_VIDEO_FRAMES)
AttributeError: 'MainWindow' object has no attribute 'TOTAL_VIDEO_FRAMES'

we need to add a condition for video modes only
@Ahmed-Gazzar

Hello, will there be an docker image or docker file that runs the app in a container?

How to save the bbox?

Got keys from plugin meta data ("xcb")
QFactoryLoader::QFactoryLoader() checking directory path "/home/zhangzs36/anaconda3/envs/DLTA-AI/bin/platforms" ...
loaded library "/home/zhangzs36/anaconda3/envs/DLTA-AI/lib/python3.8/site-packages/cv2/qt/plugins/platforms/libqxcb.so"
QObject::moveToThread: Current thread (0x27436a0) is not the object's thread (0x8671ba0).
Cannot move to target thread (0x27436a0)

Package Version

addict 2.4.0
beautifulsoup4 4.12.2
brotlipy 0.7.0
certifi 2023.5.7
cffi 1.15.1
charset-normalizer 2.0.4
click 8.1.6
colorama 0.4.6
contourpy 1.1.0
cryptography 39.0.1
cycler 0.11.0
darkdetect 0.7.1
dataclasses-json 0.5.12
DLTA-AI 1.0.2
easydict 1.10
filelock 3.12.2
filterpy 1.4.5
fonttools 4.41.0
gdown 4.7.1
idna 3.4
imageio 2.31.1
imgviz 1.7.2
importlib-metadata 6.8.0
importlib-resources 6.0.0
jeepney 0.8.0
kiwisolver 1.4.4
lap 0.4.0
lazy_loader 0.3
loguru 0.6.0
Markdown 3.4.3
markdown-it-py 3.0.0
marshmallow 3.19.0
matplotlib 3.7.2
mdurl 0.1.2
mkl-fft 1.3.6
mkl-random 1.2.2
mkl-service 2.4.0
mmcv-full 1.7.0
mmdet 2.25.2
model-index 0.1.11
mypy-extensions 1.0.0
networkx 3.1
notify-py 0.3.42
numpy 1.24.3
onemetric 0.1.2
opencv-python 4.7.0.72
openmim 0.3.2
ordered-set 4.1.0
orjson 3.8.12
packaging 23.1
pandas 2.0.3
Pillow 9.4.0
pip 23.1.2
platformdirs 3.9.1
psutil 5.9.4
pycocotools 2.0.6
pycparser 2.21
Pygments 2.15.1
pyOpenSSL 23.0.0
pyparsing 3.0.9
PyQt5 5.15.9
PyQt5-Qt5 5.15.2
PyQt5-sip 12.12.1
pyqtdarktheme 2.1.0
PySocks 1.7.1
python-dateutil 2.8.2
pytz 2023.3
PyWavelets 1.4.1
PyYAML 6.0.1
QtPy 2.3.1
requests 2.29.0
rich 13.4.2
scikit-image 0.20.0
scipy 1.9.1
seaborn 0.12.2
segment-anything 1.0
sentry-sdk 1.28.1
setuptools 67.8.0
six 1.16.0
soupsieve 2.4.1
supervision 0.3.2
tabulate 0.9.0
termcolor 2.2.0
terminaltables 3.1.10
thop 0.1.1.post2209072238
tifffile 2023.7.10
tomli 2.0.1
torch 1.13.1
torchaudio 0.13.1
torchvision 0.14.1
tqdm 4.65.0
typing_extensions 4.6.3
typing-inspect 0.9.0
tzdata 2023.3
ultralytics 8.0.61
urllib3 1.26.16
wheel 0.38.4
yapf 0.40.1
zipp 3.16.2

When a tool is left to run for a long time (tracking a 1 hour video) It crashes and gives "Killed" in the terminal. The so far tracking annotations are not saved.

问题描述如上

when running YOLOv8 it creates polygons for all thresholds, however the threshold

Error message

From the working one: mmdetection/configs/solo/solo_r50_fpn_1x_coco.py
load checkpoint from local path: mmdetection/checkpoints/solo_r50_fpn_1x_coco_20210821_035055-2290a6b8.pth
Decoding .../data/imgs\camels.jpg
Time taken to annoatate img on SOLO: 613 ms
Decoding .../data/imgs\img1.jpg
Time taken to annoatate img on SOLO: 615 ms
'dict' object has no attribute 'other_data'
Decoding .../data/imgs\img1_corrupted.jpg
Time taken to annoatate img on SOLO: 328 ms
Decoding .../data/imgs\img2.jpg
Time taken to annoatate img on SOLO: 804 ms
'dict' object has no attribute 'other_data'
Decoding .../data/imgs\img3.jpg
Time taken to annoatate img on SOLO: 526 ms
'dict' object has no attribute 'other_data'
Decoding .../data/imgs\test1_Moment1.jpg
Time taken to annoatate img on SOLO: 1543 ms
'dict' object has no attribute 'other_data'
Decoding .../data/imgs\test1_Moment2.jpg
Time taken to annoatate img on SOLO: 1250 ms
'dict' object has no attribute 'other_data'
Decoding .../data/imgs\test2_Moment1.jpg
Time taken to annoatate img on SOLO: 1140 ms
'dict' object has no attribute 'other_data'
Decoding .../data/imgs\test2_Moment2.jpg
Time taken to annoatate img on SOLO: 826 ms
'dict' object has no attribute 'other_data'

by adding dropdown menu to select tracking type (for number of frames, assigned objects, whole video)

Error message

OpenCV(4.7.0) :-1: error: (-5:Bad argument) in function 'resize'
> Overload resolution failed:
>  - src is not a numpy array, neither a scalar
>  - Expected Ptr<cv::UMat> for argument 'src'

Note: The tool doesn't crash, just no inference

After annotating some data, I tried to export the annotation in COCO format, but I am getting the error:
'NoneType' object has no attribute 'filename

version = pip 23.2 (python 3.8)

Preparing metadata (pyproject.toml) ... error
  error: subprocess-exited-with-error

  × Preparing metadata (pyproject.toml) did not run successfully.
  │ exit code: 1
  ╰─> [21 lines of output]
      <string>:4: DeprecationWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html
      Partial import of lap during the build process.
      Traceback (most recent call last):
        File "<string>", line 127, in get_numpy_status
      ModuleNotFoundError: No module named 'numpy'
      Traceback (most recent call last):
        File "c:\users\carlos\workspace\trackia-env\lib\site-packages\pip\_vendor\pyproject_hooks\_in_process\_in_process.py", line 353, in <module>
          main()
        File "c:\users\carlos\workspace\trackia-env\lib\site-packages\pip\_vendor\pyproject_hooks\_in_process\_in_process.py", line 335, in main
          json_out['return_val'] = hook(**hook_input['kwargs'])
        File "c:\users\carlos\workspace\trackia-env\lib\site-packages\pip\_vendor\pyproject_hooks\_in_process\_in_process.py", line 149, in prepare_metadata_for_build_wheel
          return hook(metadata_directory, config_settings)
        File "C:\Users\carlos\AppData\Local\Temp\pip-build-env-zmrkanjx\overlay\Lib\site-packages\setuptools\build_meta.py", line 380, in prepare_metadata_for_build_wheel
          self.run_setup()
        File "C:\Users\carlos\AppData\Local\Temp\pip-build-env-zmrkanjx\overlay\Lib\site-packages\setuptools\build_meta.py", line 487, in run_setup
          super(_BuildMetaLegacyBackend,
        File "C:\Users\carlos\AppData\Local\Temp\pip-build-env-zmrkanjx\overlay\Lib\site-packages\setuptools\build_meta.py", line 338, in run_setup
          exec(code, locals())
        File "<string>", line 236, in <module>
        File "<string>", line 220, in setup_package
      ImportError: lap requires numpy, please "pip install numpy".
      [end of output]

  note: This error originates from a subprocess, and is likely not a problem with pip.
error: metadata-generation-failed

× Encountered error while generating package metadata.
╰─> See above for output.

note: This is an issue with the package mentioned above, not pip.
hint: See above for details.

Is there an executable file for DLTA-AI that I can download?

When you run DLTA-AI and you get an error in mmcv, you need to install mmcv. If you get an error when installing mmcv," like:

you can try this:
pip install mmcv-full==1.3.17 -f https://download.openmmlab.com/mmcv/dist/11.1/torch1.10.0/index.html

Hello and thank you for sharing this great tool. If you take a look at Labelme, there is a checkbox in the file menu that gives you the option of not saving the image data with the exported JSON file. But I don't see this checkbox in this version, and the user is forced to save the image data with the JSON file. I guess saving the image data makes the saved file too big (as big as the image file) which can be challenging for those who don't have enough memory.
It would be helpful if you consider this option in the upcoming versions.

The app just loads the loading page before closing with this issue:

We can generator saved_models.json from the Model Explorer by taking to most used 3 or 4 models, they can be accessed directly from Select Segmentation model menu

like RLE, segmentation or bbox only

instead of downloading manually

Is there any way to load annotation in COCO/YOLO format and modify it? This would be very helpful.

both using Ctrl+E
I tried to modify Edit Label, but it still Ctrl+E in the Context Menu (Right Click)

e.g., when pressing on create polygon: SAM Toolbar appears and you have the option to create / edit manullay or using SAM

Got this Error:

YOLOv8n-seg summary (fused): 195 layers, 3404320 parameters, 0 gradients, 12.6 GFLOPs
Traceback (most recent call last):
  File "C:\Users\xxxx\Documents\DLTA-AI\DLTA_AI_app\labelme\widgets\canvas.py", line 675, in paintEvent
    p.translate(self.offsetToCenter())
  File "C:\Users\xxxx\Documents\DLTA-AI\DLTA_AI_app\labelme\widgets\canvas.py", line 822, in offsetToCenter
    return QtCore.QPoint(x, y)
TypeError: arguments did not match any overloaded call:
  QPoint(): too many arguments
  QPoint(xpos: int, ypos: int): argument 1 has unexpected type 'float'
  QPoint(a0: QPoint): argument 1 has unexpected type 'float'

I'm getting a lot of these Qt type errors. If I fix them by converting the float to int the app works.

I fixed the error from above for me by changing
..\DLTA-AI\DLTA_AI_app\labelme\widgets\canvas.py line 822 to

return QtCore.QPoint(int(x), int(y))

I'm working remote on the machine where the software is installed and using a Widescreen monitor. Maybe that's why the calculations produce floats.

Hope it helps,
best wishes and thanks for the tool. :)