The nature of statistical learning theory
The nature of statistical learning theory
Statistical Pattern Recognition: A Review
IEEE Transactions on Pattern Analysis and Machine Intelligence
A robust audio classification and segmentation method
MULTIMEDIA '01 Proceedings of the ninth ACM international conference on Multimedia
Pattern Classification (2nd Edition)
Pattern Classification (2nd Edition)
Automatic Analysis of Multimodal Group Actions in Meetings
IEEE Transactions on Pattern Analysis and Machine Intelligence
Robust Scene Categorization by Learning Image Statistics in Context
CVPRW '06 Proceedings of the 2006 Conference on Computer Vision and Pattern Recognition Workshop
Supervised classification for video shot segmentation
ICME '03 Proceedings of the 2003 International Conference on Multimedia and Expo - Volume 1
The Semantic Pathfinder: Using an Authoring Metaphor for Generic Multimedia Indexing
IEEE Transactions on Pattern Analysis and Machine Intelligence
The challenge problem for automated detection of 101 semantic concepts in multimedia
MULTIMEDIA '06 Proceedings of the 14th annual ACM international conference on Multimedia
A probabilistic framework for semantic video indexing, filtering,and retrieval
IEEE Transactions on Multimedia
A unified framework for semantic shot classification in sports video
IEEE Transactions on Multimedia
Exploiting redundancy in cross-channel video retrieval
Proceedings of the international workshop on Workshop on multimedia information retrieval
Foundations and Trends in Information Retrieval
Episode-constrained cross-validation in video concept retrieval
IEEE Transactions on Multimedia
Hi-index | 0.00 |
Digital video is sequential in nature. When video data is used in a semantic concept classification task, the episodes are usually summarized with shots. The shots are annotated as containing, or not containing, a certain concept resulting in a labeled dataset. These labeled shots can subsequently be used by supervised learning methods (classifiers) where they are trained to predict the absence or presence of the concept in unseen shots and episodes. The performance of such automatic classification systems is usually estimated with cross-validation. By taking random samples from the dataset for training and testing as such, part of the shots from an episode are in the training set and another part from the same episode is in the test set. Accordingly, data dependence between training and test set is introduced, resulting in too optimistic performance estimates. In this paper, we experimentally show this bias, and propose how this bias can be prevented using episode-constrained crossvalidation. Moreover, we show that a 17% higher classifier performance can be achieved by using episode constrained cross-validation for classifier parameter tuning.