Accepted in IEEE National Conference on Computer Vision, Pattern Recognition, Image Processing and Graphics (NCVPRIPG),2013

Chiranjoy Chattopadhyay and Sukhendu Das
Visualization and Perception Lab
Department of Computer Science and Engineering, Indian Institute of Technology, Madras, India

This paper presents the design of STAR (SpatioTemporal Analysis and Retrieval), an unsupervised Content Based Video Retrieval (CBVR) System. STAR's key insight and primary contribution is that it models video content using a joint spatio-temporal feature representation and retrieves videos from the database which have similar moving object and trajectories of motion. Foreground moving blobs from a moving camera video shot are extracted, along with a trajectory for camera motion compensation, to form the space-time volume (STV). The STV is processed to obtain the EMST-CSS representation, which can discriminate across different categories of videos. Performance of STAR has been evaluated qualitatively and quantitatively using precision-recall metric on benchmark video datasets having unconstrained video shots, to exhibit efficiency of STAR.

Framework

Results

*Median frame of top 9 tank ordered (row-wise) retrieved video shots

"STAR: A Content Based Video Retrieval System for Moving Camera Video Shots", Chiranjoy Chattopadhyay and Sukhendu Das; Accepted in the National Conference on Computer Vision, Pattern Recognition, Image Processing and Graphics (NCVPRIPG-2013), Jodhpur, India, 19-21 December 2013

@inproceedings{star, author={Chiranjoy Chattopadhyay and Sukhendu Das}, booktitle={{National Conference on Computer Vision, Pattern Recognition, Image Processing and Graphics (NCVPRIPG-2013)}},note={{Jodhpur, India, 19-21 December 2013, Accepted. To appear}}, title={{STAR: A Content Based Video Retrieval System for Moving Camera Video Shots. }}, year={2013}, }