Massachusetts Institute of Technology

Jason R Thornton

Michael T Chan

Daniel J Butler

Jeanette T Baran-Gale

Aaron Z Yahr

Christine M Russ

Heather Zwahlen

Ronald Duarte

Marianne A Deangelus

Benjamin A Miller

Kevin Kyung Nam

Daniel J Scarafoni

Luke Skelly

Nicholas Gallo

Zach Elko

Described herein are systems and methods for locating an object detected in a video. The system detects a bounding box at least partially around an object in a first frame at a first time in the video and a second frame in the video corresponding to a second time. The system determines whether there is no motion within the bounding box of the second frame. The system compares edge information, or color information, or intensity information associated with one or more pixels in the first frame, to edge information, or color information, or intensity information associated with one or more pixels within the bounding box. The system generates a score based on the comparison. The system further determines based on the score if the object is present in the second frame. The system also determines an estimated timeframe window of a first appearance of the object.

Forensic video exploitation and analysis tools

Systems and methods are provided for selecting an optimized data model architecture subject to resource constraints. One or more resource constraints for target deployment are identified, and random model architectures are generated from a set of model architecture production rules subject to the one or more resource constraints. Each random model architecture is defined by randomly chosen values for one or more meta parameters and one or more layer parameters. One or more of the random model architectures are adaptively refined to improve performance relative to a metric, and the refined model architecture with the best performance relative to the metric is selected.

Systems and methods for optimization of a data model network architecture for target deployment

Exemplary embodiments include a system and method for Automatic Customization of Content Filtering (ACCF). The method includes receiving a search string from a user, creating a first filter based on the search string and searching different types of content stored in different content locations based on the first filter. The search returns a first subset of results to the user based on the first filter. The method further receives an indication of relevance for each one of the results in the first subset from the user. The method dynamically creates a second filter based on the received indications of relevance for the first subset of results. Based on the second filter, a second subset of more relevant results is returned to the user.

Systems and methods for automatic customization of content filtering

Cyclical pixel-based remapping techniques are disclosed for efficiently creating and displaying a video summary. Cyclical temporal remapping at the pixel level requires less processing power and eliminates the need for object identification and tracking, allowing for dynamic adjustment of summary video parameters while a user is viewing the summary video. An operator can select an archived camera feeds or video file, specify a time compression ratio or time interval over which to review video content, specify a motion sensitivity threshold, and then view a significantly condensed summary video that depicts all activity from the source video within a much shorter timeline. Furthermore, activity co-occurrence is preserved such that any activities that co-occur in the source video will also co-occur in the summary video.

Dynamic video summarization

A computer-implemented image processing method includes accessing image data representing a plurality of images of a plurality of vehicles and reading a data model representing a predefined vehicle attribute. The method further includes processing a portion of the image data to detect a vehicle feature of at least one of the vehicles represented in at least one of the images based on the predefined vehicle attribute, and processing the portion of the image data to generate vehicle data representing the detected vehicle feature.

Assisted surveillance of vehicles-of-interest

Methods, systems and media are described for computer-assisted video surveillance. Methods may support detection of moving persons in video frames, extraction of features of the detected moving persons and identification of which detected moving persons are likely matches to a person of interest. Identification of the likely matches may be determined using an attribute-based search, and/or using a specific person-based search. The method may include using likely matches confirmed as images of the person of interest to reconstruct a path of the person of interest.

Growing community of inventors

Chelmsford, MA, United States of America

Jason R Thornton