MindTree, a global product engineering services and IT solutions company, today unveiled its real - time video stitching algorithm, which is part of its intelligent video surveillance solution. The patent - pending algorithm can stitch videos from up to four co-located cameras placed either horizontally or vertically in real time. The stitched video from multiple cameras increases in field of view and provides better situation awareness to the viewer.
MindTree’s intelligent video and provides better situation awareness to the viewer. It’s intelligent video surveillance programme is supported by the Council of Scientific & Industrial Research (CSIR) under its New Millennium Indian Technology Leadership Initiative (NMITLI).
The low-complexity video stitching algorithm requires no scene training and performs dynamic feature matching and blending. A scene overlap of at least five percent is required for the algorithm to work.
These algorithms have been optimised on the Intel and TI platforms to reduce the time of market. Typical locations where video stitching is deployed to provide greater field of view are railways stations, airports, retail stores, border and coastal surveillance points. Video stitching is also deployed in high-end automobiles and in battle tanks to enhance passenger safety by increasing the field of view for the driver.