Foveated Video Object Detection Navy STTR FY2014.A Sol No.: Navy STTR FY2014.A Topic No.: N14A-T008 Topic Title: Foveated Video Object Detection Proposal No.: N14A-008-0092 Firm: Mayachitra, Inc. 5266 Hollister Avenue, Suite 229 Santa Barbara, California 93111 Contact: Jelena Tesic Phone: (646) 379-6042 Abstract: Humans are still unsurpassed in their ability to search for objects in visual scenes. On the other hand, object detection continues to be one of the hot areas of computer vision, and it attests both to the fundamental importance of the problem and the fact that existing state-of-the-art algorithms are still shy of performing well in practice. We propose to utilizes human vision modeling of a foveated view along with saccadic movements of the eye to quickly and accurately find and track objects of interest. We will develop a framework that combines state-of-the-art commercial grade video object detection system with foveated visual field and saccade fixation point modeling for better and faster object detection in UAS video feeds. Benefits: 1.Total data exploitation:precise and fast object detection on-board of sensor stations allows for better/more detailed UAS targets/object of interest intelligence feed. 2. Scalable processing The system will utilize foviated visual field view, saccade type selection of fixation points, video objectness and compressive sensing for more relevant data to be collected, transmitted faster, and for more alerts to be automatically generated. 3. Context-aware motion imagery analysis: The tracking and object recognition algorithms will model the spatial and temporal evolution of objects/activities to achieve long-term tracking and recognition with high-accuracy in monitored wide areas that are high-density and cluttered. Return