A Novel Learning-Based Framework for Detecting Interesting Events in Soccer Videos
Authors:
- Nisha JainNJNisha Jain
- Santanu ChaudhurySCSantanu Chaudhury
- Sumantra Dutta RoySRSumantra Dutta Roy
- Prasenjit MukherjeePMPrasenjit Mukherjee
- Krishanu SealKSKrishanu Seal
- Kumar TalluriKTKumar Talluri
- 1 December 2008
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 119-125
Abstract
We present a novel learning-based framework for detecting interesting events in soccer videos. The input to the system is a raw soccer video. We have learning at three levels-learning to detect interesting low-level features from image and video data using support vector machines (hereafter, SVMs), and a hierarchical conditional random field- (hereafter, CRF-) based methodology to learn the dependencies of mid-level features and their relation with the low level features, and high level decisions (dasiainteresting eventspsila) and their relation with the mid-level features: all on the basis of training video data. Descriptors are spatio-temporal in nature - they can be associated with a region in an image or a set of frames. Temporal patterns of descriptors characterise an event. We apply this framework to parse soccer videos into Interesting (a goal or a goal miss) and Non-Interesting videos. We present results of numerous experiments in support of the proposed strategy.Keywords
This publication has 13 references indexed in Scilit:
- Large-scale multimodal semantic concept detection for consumer video2007
- Unsupervised content-based indexing of sports video2007
- Hierarchical Conditional Random Fields for GPS-Based Activity Recognition2007
- Learning, detection and representation of multi-agent events in videosArtificial Intelligence, 2007
- Video abstraction: A systematic review and classificationACM Transactions on Multimedia Computing, Communications, and Applications, 2007
- Automatic annotation and semantic retrieval of video sequences using multimedia ontologies2006
- Fusion of AV features and external information sources for event detection in team sports videoACM Transactions on Multimedia Computing, Communications, and Applications, 2006
- Robust Real-Time Face DetectionInternational Journal of Computer Vision, 2004
- A mid-level representation framework for semantic sports video analysis2003
- Highlight scene extraction in real time from baseball live video2003