video_to_image_seq.py -- A Python function to convert a video to a sequence of images. Depends on OpenCV. See the original and an application here: https://github.com/talhaahussain/grappling-pose-identification/

video_to_image_seq.py

import cv2

def split_video_into_frames(filename):
  frames = []
  cap = cv2.VideoCapture(filename)
  fps = cap.get(cv2.CAP_PROP_FPS)

  success = 1
  while success:
    success, frame = cap.read()
    if success:
      frames.append(frame)

  cap.release()
  return frames, fps