Hi Support Forum,
I'm interested in using computer vision services for my videos (for classification, object detection, etc). One way to start is by pulling out frames, and sending the images to a service. I saw that I can transform a video into a gif, effectively sampling video frames every x seconds here: https://cloudinary.com/documentation/videos_to_animated_images. But I'm cheap! is there a smarter way to sample frames that represent different shots in the video?
Post is closed for comments.