Abstract: A video information extraction system includes a memory to store a video; a textual information extraction module to obtain information about terms, entities, relations, and events from a ground truth caption corresponding to the video; and a video captioning module including an encoder (i) to receive the information about the terms, entities, relations, and events from the textual information extraction module, and (ii) to extract video features from the video; and a decoder to generate a text caption based on the extracted video features.

Abstract: A method is provided for accurately determining the registration for a moving vehicle over a number of frames so that the vehicle can be super-resolved. Instead of causing artifacts in a super-resolved image, the moving vehicle can be specifically registered and super-resolved individually. This method is very accurate, as it uses a mathematical model that captures motion with a minimal number of parameters and uses all available image information to solve for those parameters. Methods are provided that implement the vehicle registration algorithm and super-resolve moving vehicles using the resulting vehicle registration. One advantage of this system is that better images of moving vehicles can be created without requiring costly new aerial surveillance equipment.

Abstract: A system recognizes events. The system includes a sequence of continuous vectors and a sequence of binarized vectors. The sequence of continuous vectors represents spatial-dynamic relationships of objects in a predetermined recognition area. The sequence of binarized vectors is derived from the sequence of continuous vectors by utilizing thresholds for determining binary values for each spatial-dynamic relationship. The sequence of binarized vectors indicates whether an event has occurred.

Abstract: A system recognizes events. The system includes a sequence of continuous vectors and a sequence of binarized vectors. The sequence of continuous vectors represents spatial-dynamic relationships of objects in a predetermined recognition area. The sequence of binarized vectors is derived from the sequence of continuous vectors by utilizing thresholds for determining binary values for each spatial-dynamic relationship. The sequence of binarized vectors indicates whether an event has occurred.