Date of Publication :7th May 2016
Abstract: Most of the object detectors performance is degraded due to change in illumination, variant background and camera position. A method to enhance the detection performance of an offline generic detector is proposed in the paper. In this approach, all the detections are represented in Bag of Word fashion considering super pixels as its feature of classification, combining super pixels extraction and bag of word improves the object detection of a generic offline detector, object shape extraction from its background is segmented using graph cut algorithm. In standard, proposed approach takes the detection bounding box generated by a generic detector as input and improves the detection with better average precision. Bounding boxes are reduced with the objects shapes giving better performance using graph cut algorithm.
Reference :
-
[1] P. Felzenszwalb, R. Girshick, D. McAllester, and D. Ramanan. “Object detection withdiscriminatively trained part based models”, In PAMI, 2010.
[2] Navneet Dalal and Bill Triggs “Histograms of Oriented Gradients for Human Detection”, in CVPR 2005.
[3] V. Nair and J. Clark. “An unsupervised, online learning framework for moving object detection”, In CVPR, 2004
[4] A. Levin, P. Viola, and Y. Freund. “Unsupervised improvement of visual detectors using co-training”, In CVPR, 2003.
[5] R. Achanta, A. Shaji, P. F. Kevin Smith, Aurelien Lucchi, and S. Susstrunk. “Slic superpixels”, In EPFL Technical Report, 2010.
[6]Y. Boykov and Jolly. “Interactive graph cuts for optimal boundary and region segmentation of objects in nd images”. In ICCV, 2001.
[7] P.ViolaandM.Jones. “Rapid object detection using a boosted cascade of simple features”, In CVPR, 2001.
[8]V. Nair and J. Clark. An unsupervised, online learning framework for moving object detection. In CVPR, 2004
[9] A. R. Zamir, A. Dehghan, and M. Shah. Gmcp-tracker “Global multi-object tracking using generalized minimum clique graphs”, In ECCV, 2012.
[10] G.Shu, A. Dehghan, O.Oreifej, E.Hand, and M.Shah. “Part based multiple-person tracking with partial occlusion handling”, In CVPR, 2012.
[11] B. Benfold and I. Reid. “Stable multi-target tracking in real time surveillance video”, In CVPR, 2011.
[12] H. Celik, A. Hanjalic, and E. A. Hendriks. “Unsupervised and simultaneous training of multiple object detectors from unlabeled surveillance video”, In Computer Vision and Image Understanding, 2009.
[13] Guang Shu, Afshin Dehghan, Mubarak Shah “Improving an Object Detector and Extracting Regions using Superpixels”,In CVPR,2013