Smartphones and other digital cameras enable users to capture photos that include several frames, including some prior to the instant of the user selecting the capture button and some after. This allows motion within the scene to be captured as well as for post-capture selection of a suitable still image, e.g., an image where the subjects are in the correct pose and/or are looking at the camera. However, this is wasteful of storage space when the captured image is of a still object. This disclosure describes techniques to automatically detect such multi-frame photos and enable users to remove excess frames and obtain a standard still image file that is of a smaller size. Automatic detection can be performed using two separate classifiers that employ suitable computer vision techniques to detect photos that contain static content and that do not depict living beings.

Creative Commons License

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.