pythonnumpyopencvgame-automation

File Unplayable when using VideoWriter with OpenCV


The .mp4 file that gets created by the VideoWriter gets an error when I try to play it that says "This file isn't playable. That might be because the file type is unsupported, the file extension is incorrect, or the file is corrupt"

I tried moving my VideoWriter setup to my other file just to see and I had the same result with that. Im rather new to OpenCV and Numpy so Im not really sure how I should go about this.

This is the code in my main.py file:

import cv2 as cv
import numpy as np
from time import time
import datetime

from windowcapture import WindowCapture

wincap = WindowCapture('Geometry Dash')


time_stamp = datetime.datetime.now().strftime('%Y-%m-%d %H-%M-%S')
fourcc = cv.VideoWriter_fourcc('m', 'p', '4', 'v')
file_name = f'{time_stamp}.mp4'
captured_video = cv.VideoWriter(file_name, fourcc, 60.0, (800, 600))

loop_time = time()
while True:

    screenshot = wincap.get_screenshot()

    # prints the video fps
    print('FPS {:.2f}'.format(1 / (time() - loop_time)))
    loop_time = time()
    
    cv.imshow('Screen Capture', screenshot)
    captured_video.write(screenshot)


    # press 'q' with output window focused to close window
    # waits 1ms every loop to process if q is pressed
    if cv.waitKey(1) == ord('q'):
        cv.destroyAllWindows()
        break


print('q pressed // window closing')

This is the code in my windowcapture.py file:

import numpy as np
import win32gui, win32ui, win32con


class WindowCapture:

    # properties
    w = 0
    h = 0
    hwnd = None
    cropped_x = 0
    cropped_y = 0
    offset_x = 0
    offset_y = 0

    # constructor
    def __init__(self, window_name):

        self.hwnd = win32gui.FindWindow(None, window_name)
        if not self.hwnd:
            raise Exception('Window not found "{}"'.format(window_name))
        
        # get window size
        window_rect = win32gui.GetWindowRect(self.hwnd)
        self.w = window_rect[2] - window_rect[0]
        self.h = window_rect[3] - window_rect[1]

        # account for window border and titlebar and cut them off
        border_pixels = 8
        titlebar_pixels = 30
        self.w = self.w - (border_pixels * 2)
        self.h = self.h - titlebar_pixels - border_pixels
        self.cropped_x = border_pixels
        self.cropped_y = titlebar_pixels

        # set the cropped coordinates offset so we can translate screenshot
        # images into actual screen positions
        self.offset_x = window_rect[0] + self.cropped_x
        self.offset_y = window_rect[1] + self.cropped_y

    def get_screenshot(self):
        
        # get the window image data
        wDC = win32gui.GetWindowDC(self.hwnd)
        dcObj = win32ui.CreateDCFromHandle(wDC)
        cDC = dcObj.CreateCompatibleDC()
        dataBitMap = win32ui.CreateBitmap()
        dataBitMap.CreateCompatibleBitmap(dcObj, self.w, self.h)
        cDC.SelectObject(dataBitMap)
        cDC.BitBlt((0, 0), (self.w, self.h), dcObj, (self.cropped_x, self.cropped_y), win32con.SRCCOPY)

        # saves the screenshot
        #dataBitMap.SaveBitmapFile(cDC, 'debug.bmp')
        
        signedIntsArray = dataBitMap.GetBitmapBits(True)
        img = np.frombuffer(signedIntsArray, dtype='uint8')
        img.shape = (self.h, self.w, 4)

        # Free Resources
        dcObj.DeleteDC()
        cDC.DeleteDC()
        win32gui.ReleaseDC(self.hwnd, wDC)
        win32gui.DeleteObject(dataBitMap.GetHandle())

        # make image C_CONTIGUOUS to avoid errors like:
        #   File ... in draw_rectangles
        #   TypeError: an integer is required (got type tuple)
        # github discussion:
        # https://github.com/opencv/opencv/issues/14866#issuecomment-580207109 
        img = np.array(img)
        return img
    
    # find the name of the window you're interested in.
    # once you have it, update window_capture()
    # https://stackoverflow.com/questions/55547940/how-to-get-a-list-of-the-name-of-every-open-window
    def list_window_names(self):
        def winEnumHandler(hwnd, ctx):
            if win32gui.IsWindowVisible(hwnd):
                print(hex(hwnd), win32gui.GetWindowText(hwnd))
        win32gui.EnumWindows(winEnumHandler, None)

    #list_window_names()

    # translate a pixel position on a screenshot image to a pixel position on the screen.
    # pos = (x,y)
    # WARNING: if you move the window being captured after execution is started, this will
    # return incorrect coordinates, because the window position is only calculated in
    # the __init__ constructor.
    def get_screen_position(self, pos):
        return (pos[0] + self.offset_x, pos[1] + self.offset_y)
'''
I have no problem viewing my video when I have my game open it plays perfectly fine in the window that pops up from opencv but when I try to save this video to mp4 the video does not work.

Solution

  • This may be more of a temporary answer to my question because this doesn't really feel like the "correct" way of solving this but this is what I did.

    In my main.py I changed

    time_stamp = datetime.datetime.now().strftime('%Y-%m-%d %H-%M-%S')
    fourcc = cv.VideoWriter_fourcc('m', 'p', '4', 'v')
    file_name = f'{time_stamp}.mp4'
    captured_video = cv.VideoWriter(file_name, fourcc, 60.0, (800, 600))
    

    Ive changed this code to

    time_stamp = datetime.datetime.now().strftime('%Y-%m-%d %H-%M-%S')
    fourcc = cv.VideoWriter_fourcc(*'mp4v')
    file_name = f'{time_stamp}.mp4'
    captured_video = cv.VideoWriter(file_name, fourcc, 60.0, (1600,900))
    

    and then in my windowcapture.py file I changed

    img = np.array(img)
    return img
    

    Ive now changed that to

    img_np = np.array(img)
    img_bgr = cv.cvtColor(img_np, cv.COLOR_RGB2BGR)
    img_rgb = cv.cvtColor(img_bgr, cv.COLOR_BGR2RGB)
    return img_rgb
    

    The rest of the code I've left alone and I now get my mp4 files to play. The first change that I noticed from the testing I did was I needed to set my VideoWriter to the correct screensize as I had it in 800, 600 when the window was 1600, 900. The second change was the color conversions If I only convert to BGR the code works but obviously I don't want it in BGR I want it in RGB so I had to convert back, but if I don't do at least one of these conversions the mp4 will not play.