I have a working Yolov8 model that detects playing cards in groups or individually. It was trained on images and I would like it to work with video clips. I have exported it to a tflite format and it works for images but fails with video clips. Can Frigate work with models trained on images or do I need to train a new model on video clips?