WMA1 - Allows to find optimal parameters for each frame of video to make good tracking of object. Also allows to generate video with tracking.

WMA2 - Looking for coins on the tray. Recognizes if the coin inside or outside the tray and recognizes it's value. Also counts sum of all coins.

WMA3 - Using photoes makes keypoints of objet and can track this object on the video. Also when enough keypoints shows mask.

WMA4 - Using CNN recognizes celebrities which was in train data, also can separate celebrities from other people.

WMA5 - Compare YOLO with pre-trained weights and completely new weights

WMA6 - Using CV2 to recognize face and make glasses mask