yolov5windowdetect 0.11

Capture window - run yolov5 - show results

pip install yolov5windowdetect

Update 2022/10/25

Run multiple detection models

winca = (

    Yolov5WindowDetect(

        pt_file=r"C:\Users\Gamer\anaconda3\envs\dfdir\yolov5\runs\train\playerbutton7\weights\best.pt",

        repo_or_dir="./yolov5",

        model="custom",

        source="local",

    )

    .add_models(

        pt_file=r"C:\Users\Gamer\anaconda3\envs\dfdir\yolov5\runs\train\playerbutton6\weights\best.pt",

        repo_or_dir="./yolov5",

        model="custom",

        source="local",

    )

    .get_hwnd_window(window_title_regex=r"[Bb]lue[sS]tacks.*")

    .take_screenshot_and_run_yolov(

        confidence_thresh=0.3,

        show_results=True,

        quit_key="q",

        sleep_time=0.04,

        rununtilstopped=True,

    )

)

from yolov5windowdetect import Yolov5WindowDetect

winca = (

    Yolov5WindowDetect(

        pt_file=r"C:\Users\.....\best.pt",

        repo_or_dir="./yolov5",

        model="custom",

        source="local",

    )

    .get_hwnd_window(window_title_regex=r"[Bb]lue[Ss]tacks.*")

    .take_screenshot_and_run_yolov(

        confidence_thresh=0.3,

        show_results=True,

        quit_key="q",

        sleep_time=0.04,

        rununtilstopped=True,

    )

)



# window capture -> don't show results, only results as df

winca.get_hwnd_window(

    window_title_regex=r"[Bb]lue[Ss]tacks.*"

).take_screenshot_and_run_yolov(

    confidence_thresh=0.3, show_results=False, rununtilstopped=False,

).get_results_as_df()





# adb capture

winca.get_adb_window(

    adb_path=r"C:\ProgramData\adb\adb.exe", adb_serial="localhost:5555",

).take_screenshot_and_run_yolov(

    confidence_thresh=0.3, show_results=False, rununtilstopped=False,

).get_results_as_df()

Example

  <a href="https://www.youtube.com/watch?v=-jXqL39Tf5w">

     <img src="https://img.youtube.com/vi/-jXqL39Tf5w/0.jpg" style="width:100%;">

  </a>