Abstract:Content-based video search reranking can be regarded as a process that uses visual content to recover the "true" ranking list from the noisy one generated based on textual information. This paper explicitly formulates this problem in the Bayesian framework, i.e., maximizing the ranking score consistency among visually similar video shots while minimizing the ranking distance, which represents the disagreement between the objective ranking list and the initial text-based. Different from existing point-wise ranking distance measures, which compute the distance in terms of the individual scores, two new methods are proposed in this paper to measure the ranking distance based on the disagreement in terms of pair-wise orders. Specifically, hinge distance penalizes the pairs with reversed order according to the degree of the reverse, while preference strength distance further considers the preference degree. By incorporating the proposed distances into the optimization objective, two reranking methods are developed which are solved using quadratic programming and matrix computation respectively. Evaluation on TRECVID video search benchmark shows that the performance improvement up to 21% on TRECVID 2006 and 61.11% on TRECVID 2007 are achieved relative to text search baseline.

Video Retrieval Relevance Feedback Algorithm Based on Bayesian Learning

An Image Retrieval Relevance Feedback Algorithm Based on the Bayesian Classifier

Adaptive relevance feedback based on Bayesian inference for image retrieval

Relevance Feedback in Content-Based Image Retrieval: Bayesian Framework, Feature Subspaces, and Progressive Learning

Bayesian Video Search Reranking.

RELEVANCE FEEDBACK OF VIDEO RETRIEVAL

Relevance Feedback For Keyword And Visual Feature-Based Image Retrieval

Support Vector Machine Based Relevance Feedback Algorithm in Image Retrieval

An Effective and Simple Relevance Feedback Algorithm for Image Retrieval

Review of Relevance Feedback Algorithms in Content-based Image Retrieval

A Survey on Relevance Feedback Techniques in Content-based Medical Image Retrieval

Novel Algorithm in Relevance Feedback in Region-Based Image Retrieval System.

Fully utilize feedbacks: language model based relevance feedback in information retrieval

Non-negative Matrix Factorization Based Relevance Feedback Algorithm in Image Retrieval

Support Vector Machine Learning for Image Retrieval

A Novel SVM Based Relevance Feedback Algorithm in Image Retrieval

Relevance feedback for real-world human action retrieval

Relevance Feedback Learning with Feature Selection in Region-Based Image Retrieval

Learning based relevance feedback in image retrieval

A Powerful Relevance Feedback Mechanism for Content-Based 3D Model Retrieval

Neighborhood Preserving-based Relevance Feedback Algorithm in Image Retrieval