Automatic partitioning of full-motion video
Multimedia Systems
MULTIMEDIA '94 Proceedings of the second ACM international conference on Multimedia
A feature-based algorithm for detecting and classifying scene breaks
Proceedings of the third ACM international conference on Multimedia
IEEE Transactions on Pattern Analysis and Machine Intelligence
Performance Characterization and Comparison of Video Indexing Algorithms
CVPR '98 Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition
A feature-based algorithm for detecting and classifying production effects
Multimedia Systems
On the detection and recognition of television commercials
ICMCS '97 Proceedings of the 1997 International Conference on Multimedia Computing and Systems
| Hi-index | 0.00 |
Numerous algorithms have been proposed in the last decade, for detecting low level features and camera motion in video sequences. Along with different camera movements such as pan, tilt, and zoom, there may be some jitter movements due to hand shaking. Differentiating these jitter movements from actual camera motions is a very challenging task and is the main cause of false detections in most of the processes. Comparison of the work of different research groups working in this area can be seen from TRECVID 2005 & 2006 results. Low Precision and Recall of different methods is a clear evident that no method is suitable for all types of video to detect different camera movements. Further many methods are either computation intensive or too much false detection reduces their efficiency. A method based on region matching using area correlation is proposed to differentiate the different camera movements. Different regions are classified based on their statistical similarity in each frame of video. For each region in two consecutive frames, area correlation function is calculated. This area correlation energy function is used to detect the similarity among the regions. Weighted sum of correlation function in all the mismatched regions and location of these are used to classify the camera movements in video into different types such as pan, tilt, zoom etc. This method is tested on different videos and reasonable performance is achieved. It can also be used as pre-classifier in prior to video transition detection methods to reduce their false detections.