Author : Salini. S 1
Date of Publication :7th March 2016
Abstract: Collective activity recognizing is the action of collecting the individual person’s activity and it is distinguished as an atomic activity in isolation. In this paper, our work is based on the object detection and object recognition (ie) tracking is done for a given input video frame. The collective activity computes the class-specific person to person interaction patterns. The multiinteraction response proposed a activity – specific pattern for each interaction at the same time Naive bayes classifier is used to track the object and the tracked result is feed back into the recognition part to find category of an object and it is the final result.
Reference :
-
- M. R. Amer and S. Todorovic. A chains model for localizing partici- pants of group activities in videos. In Proc. IEEE Int. Conf. Comput. Vis. 2011 Nov, 786– 793.
- M. R. Amer, S. Todorovic, A. Fern, and S.-C. Zhu. Monte Carlo tree search for scheduling activity recognition. In Proc. IEEE Int. Conf. Comput. 2013 Dec. 1353–1360.
- S.-H. Bae and K.-J. Yoon. Robust online multiobject tracking with data association and track management. IEEE Trans. Image Process. 2014 jul. 23(7), 2820– 2833.
- W. Choi and S. Savarese. A unified framework for multi-target tracking and collective activity recognition. In Proc. 12th Eur. Conf. Comput. 2012, 4, 215–230
- SERBY, D., KOLLER-MEIER, S., AND GOOL, L. V. 2004. Probabilistic object tracking using multiple features. In IEEE International Conference of Pattern Recognition (ICPR). 184-187.
- Yilmaz, a., li, x., and shah, m. 2004. Contour based object tracking with occlusion handling in video acquired using mobile cameras. IEEE Trans. Patt. Analy. Mach. Intell. 26, 11, 1531–1536
- Silveira, g. and malis, e. 2007. real-time visual tracking under arbitrary illumination changes. ieee international conference on computer vision and pattern recognition, 1–6. 6.
- WU, Y. AND FAN, J. 2009. Contextual flow. IEEE International Conference on Computer Vision and Pattern Recognition, 33–40.
- Learning person–person interaction in collective activity recognition, xiaobin chang, wei-shi zheng, and jianguo zhang, ieee transactions on image processing, vol. 24, no. 6, june 2015.
- “Multiple Object Tracking and Segmentation in Video Sequences” .K. Meenatchi PG Scholar, Dept. of ECE CMS College of Engineering Namakkal, India , P. Subhashini Assistant Professor, Dept. of ECE CMS College of Engineering Namakkal, India
- “Cost-Sensitive Top-down/Bottom-up Inference for Multiscale Activity Recognition”Mohamed R. Amer1, Dan Xie2, Mingtian Zhao2, Sinisa Todorovic1, and Song-Chun Zhu21Oregon State University, Corvallis, oregonamerm,sinisa}@onid.orst.edu 2University of California, Los Angeles, California {xiedan,mtzhao,sczhu}@ucla.edu.
- “Learning Latent Constituents for Recognition of Group Activities in Video”Borislav Antic and Bj¨ornOmmerHCI& IWR, University of Heidelberg, Germany borislav.antic@iwr.uni-heidelberg.de, ommer@uni-heidelberg.de
- Collective Activity Dataset. [Online]. Available:http://http://www.eecs. umich.edu/vision/activitydataset.html, accessed Sep. 2013.
- Dataset: A Unified Framework for Multi-Target Tracking and Collective Activity Recognition. [Online]. Available: http://wwwpersonal