BEGIN:VCALENDAR VERSION:2.0 PRODID:-//wp-events-plugin.com//7.2.3.1//EN TZID:Asia/Kolkata X-WR-TIMEZONE:Asia/Kolkata BEGIN:VEVENT UID:120@cds.iisc.ac.in DTSTART;TZID=Asia/Kolkata:20250404T160000 DTEND;TZID=Asia/Kolkata:20250404T170000 DTSTAMP:20250403T162844Z URL:https://cds.iisc.ac.in/events/seminar-cds-102-april-04th-0400-catch-me -if-you-can-a-novel-task-for-the-detection-of-covert-geo-locations-cgl-fro m-a-single-image-view/ SUMMARY:{Seminar} @ CDS: #102\, April 04th\, 04:00: "Catch Me if You Can: A Novel Task for the Detection of Covert Geo-Locations (CGL) from a single image view. " DESCRIPTION:Department of Computational and Data Sciences\nDepartment Semin ar\n\n\n\nSpeaker : Prof. Sukhendu Das\, Dept. of Computer Science and Eng g.\, IIT Madras\nTitle : "Catch Me if You Can: A Novel Task for the Detect ion of Covert Geo-Locations (CGL) from a single image view."\nDate &\; Time : April 04\, 2025 (Friday)\, 04:00 PM\nVenue : # 102\, CDS Seminar Ha ll\n\n\n\nABSTRACT\n\nMost of the visual scene understanding tasks in the field of computer vision involve identification of the objects present in the scene. Image regions like hideouts\, corners\, bends\, turns\, and oth er obscured regions of the scene also contain crucial information for spec ific surveillance tasks. In this work\, we provide an intelligent visual a id for identification of such locations in an image\, which either pose an imminent threat or act as target zones for further investigation to ident ify any concealed objects.\n\nCovert places for hiding behind an occluding object\, are concealed locations which are not usually detectable from th e viewpoint (camera). Identification of such regions would require knowled ge about the 3D boundaries of obscuring items (like pillars\, doors\, larg e furniture\, wall bends\, endings etc.)\, spatial location and position w ith respect to the background and other neighboring regions of the scene. To advance towards the goal of identification of such regions\, we propose a novel task termed Covert GeoLocation (CGL) Detection. CGL detection fin ds applications in military counter-insurgency operations and intelligent scene surveillance with path planning for a robot. Given an input RGB imag e\, the goal is to identify all CGLs (potential hideouts) in that image. S ince it is not possible to classify any region of an image as a hideout (C GL) without looking at its surroundings\, CGL detection would require cont ext-aware detection and understanding of the complex 3D spatial relationsh ips between boundaries of occluding objects and their surroundings. We als o highlight the importance of extracting depth information for CGL detecti on\, which the traditional (object) detection approaches often do not cons ider. Our proposed method successfully extracts relevant depth features fr om only a single RGB image as input and quantitatively yields significant improvement over existing object detection and segmentation models adapted and trained for CGL detection. We also introduce a novel hand-annotated C GL detection dataset containing ~1.5K real-world images depicting CGLs in diverse indoor environments.\n\nIndex Terms: CGL detection\, hideouts\, lo cation detection\, depth perception\, visual scene understanding\, deep le arning.\n\nBIO: Prof. Sukhendu Das is currently employed as a Professor in the Dept. of Computer Science and Engg.\, IIT Madras\, Chennai\, India. H e completed his B.Tech degree from IIT Kharagpur in the Deptt. Of Electric al Engg. in 1985 and M. Tech Degree in the area of Computer Technology fro m IIT Delhi in 1987. He then obtained his Ph.D degree from IIT Kharagpur i n 1993. His current areas of research interests are: Visual Perception\, C omputer Vision: Digital Image Processing and Pattern Recognition\, Compute r Graphics\, Machine Learning and Computational Brain modeling.\n\nProf. S ukhendu Das has been a faculty of the Deptt. of CS&\;E\, IIT Madras\, I NDIA since 1989\, and awarded the CSR changemaker award by IIT Madras in 2 023. He has guided about 125 students. He has published more than 200 tech nical papers (with ~2200 citations – src: google scholar) in internation al and national journals and conferences. He has completed about 25 Sponso red projects and consultancy assignments. He is a senior member of IEEE\, and has received six (6) best papers and a best design contest award. Thre e of his MS students have recently received best MS thesis awards in the D eptt. of CS&\;E\, IIT Madras.\n\nHost Faculty: Prof. R. Venkatesh Babu\ n\n\n\nALL ARE WELCOME CATEGORIES:Events,Talks END:VEVENT BEGIN:VTIMEZONE TZID:Asia/Kolkata X-LIC-LOCATION:Asia/Kolkata BEGIN:STANDARD DTSTART:20240404T160000 TZOFFSETFROM:+0530 TZOFFSETTO:+0530 TZNAME:IST END:STANDARD END:VTIMEZONE END:VCALENDAR