Adding Inference Script to produce video sample

README.md

@@ -77,6 +77,22 @@ while True:
         break
 ```
 
+### Video Inference
+
+```shell
+usage: example_video.py [-h] --vid VID [--output OUTPUT] [--dont_show]
+
+GazeTracking Video Inference.
+
+optional arguments:
+  -h, --help       show this help message and exit
+  --vid VID        path to video
+  --output OUTPUT  inference video name. Only support .avi extension due to
+                   OpenCV 3.4
+  --dont_show      hide imshow window
+```
+
+
 ## Documentation
 
 In the following examples, `gaze` refers to an instance of the `GazeTracking` class.
@@ -166,6 +182,7 @@ Returns the main frame with pupils highlighted.
 Your suggestions, bugs reports and pull requests are welcome and appreciated. You can also starring ⭐️ the project!
 
 If the detection of your pupils is not completely optimal, you can send me a video sample of you looking in different directions. I would use it to improve the algorithm.
+* You can use `example_video.py` to produce a video sample.
 
 ## Licensing
 

example.py

@@ -11,7 +11,11 @@
 
 while True:
     # We get a new frame from the webcam
-    _, frame = webcam.read()
+    ret, frame = webcam.read()
+
+    # Quit the program if no webcam is found
+    if ret == False:
+        break
 
     # We send this frame to GazeTracking to analyze it
     gaze.refresh(frame)

example_video.py

@@ -0,0 +1,98 @@
+"""
+Demonstration of the GazeTracking library for video inference.
+Check the README.md for complete documentation.
+"""
+
+import cv2
+from gaze_tracking import GazeTracking
+import argparse
+import os
+
+# Parser
+def parser():
+    parser = argparse.ArgumentParser(description="GazeTracking Video Inference.")
+    parser.add_argument("--vid", type=str, required=True, help="path to video")
+    parser.add_argument("--output", type=str, default="result.avi", help="inference video name. Only support .avi extension due to OpenCV 3.4")
+    parser.add_argument("--dont_show", action='store_true', help="hide imshow window")
+    return parser.parse_args()
+
+def check_arguments_errors(args):
+    if not (os.path.isfile(args.vid)):
+        raise(ValueError("Invalid video path {}".format(os.path.abspath(args.video))))
+
+# Save Video
+def save_video(input_video, output_video, size):
+    fourcc = cv2.VideoWriter_fourcc(*"MJPG")
+    fps = int(input_video.get(cv2.CAP_PROP_FPS))
+    video = cv2.VideoWriter(output_video, fourcc, fps, size)
+    return video
+
+# Main
+def capture_main():
+    cap = cv2.VideoCapture(args.vid)
+
+    # Getting width and heights
+    width = int(cap.get(cv2.CAP_PROP_FRAME_WIDTH))
+    height = int(cap.get(cv2.CAP_PROP_FRAME_HEIGHT))
+
+    #Video Saver
+    video = save_video(cap, args.output, (width, height))
+
+    while cap.isOpened():
+        ret, frame = cap.read()
+        if not ret:
+            break
+
+        if frame is not None:
+
+            #Process Inference
+            image = inference(frame)
+
+            if not args.dont_show:
+                cv2.imshow('Inference', image)
+
+            video.write(image)
+
+            if cv2.waitKey(1) == 27:
+                break    
+    cap.release()
+    video.release()
+    cv2.destroyAllWindows()
+
+# Drawing
+def inference(frame):
+    # We send this frame to GazeTracking to analyze it
+    gaze.refresh(frame)
+
+    frame = gaze.annotated_frame()
+    text = ""
+
+    if gaze.is_blinking():
+        text = "Blinking"
+    elif gaze.is_right():
+        text = "Looking right"
+    elif gaze.is_left():
+        text = "Looking left"
+    elif gaze.is_center():
+        text = "Looking center"
+
+    frame = cv2.putText(frame, text, (int(frame.shape[1]*0.01), int(frame.shape[0]*0.05)), cv2.FONT_HERSHEY_DUPLEX, 1, (147, 58, 31), 2)
+
+    left_pupil = gaze.pupil_left_coords()
+    right_pupil = gaze.pupil_right_coords()
+    frame = cv2.putText(frame, "Left pupil:  " + str(left_pupil), (int(frame.shape[1]*0.01), int(frame.shape[0]*0.1)), cv2.FONT_HERSHEY_DUPLEX, 0.8, (147, 58, 31), 1)
+    frame = cv2.putText(frame, "Right pupil: " + str(right_pupil), (int(frame.shape[1]*0.01), int(frame.shape[0]*0.15)), cv2.FONT_HERSHEY_DUPLEX, 0.8, (147, 58, 31), 1)
+    return frame
+
+
+if __name__ == '__main__':
+    args = parser()
+    check_arguments_errors(args)
+
+    gaze = GazeTracking()
+
+    print('='*30)
+    print('Input Video Loaded: '+str(args.vid))
+    capture_main()
+    print('Inference Video Saved to: '+str(args.output))
+    print('='*30)