Towards effective indexing for very large video sequence database

With rapid advances in video processing technologies and ever fast increments in network bandwidth, the popularity of video content publishing and sharing has made similarity search an indispensable operation to retrieve videos of user interests. The video similarity is usually measured by the percentage of similar frames shared by two video sequences, and each frame is typically represented as a high-dimensional feature vector. Unfortunately, high complexity of video content has posed the following major challenges for fast retrieval: (a) effective and compact video representations, (b) efficient similarity measurements, and (c) efficient indexing on the compact representations. In this paper, we propose a number of methods to achieve fast similarity search for very large video database. First, each video sequence is summarized into a small number of clusters, each of which contains similar frames and is represented by a novel compact model called Video Triplet (ViTri). ViTri models a cluster as a tightly bounded hypersphere described by its position, radius, and density. The ViTri similarity is measured by the volume of intersection between two hyperspheres multiplying the minimal density, i.e., the estimated number of similar frames shared by two clusters. The total number of similar frames is then estimated to derive the overall similarity between two video sequences. Hence the time complexity of video similarity measure can be reduced greatly. To further reduce the number of similarity computations on ViTris, we introduce a new one dimensional transformation technique which rotates and shifts the original axis system using PCA in such a way that the original inter-distance between two high-dimensional vectors can be maximally retained after mapping. An efficient B+-tree is then built on the transformed one dimensional values of ViTris' positions. Such a transformation enables B+-tree to achieve its optimal performance by quickly filtering a large portion of non-similar ViTris. Our extensive experiments on real large video datasets prove the effectiveness of our proposals that outperform existing methods significantly.

3D protein structure matching by patch signatures

For determining functionality dependencies between two proteins, both represented as 3D structures, it is an essential condition that they have one or more matching structural regions called patches. As 3D structures for proteins are large, complex and constantly evolving, it is computationally expensive and very time-consuming to identify possible locations and sizes of patches for a given protein against a large protein database. In this paper, we address a vector space based representation for protein structures, where a patch is formed by the vectors within the region. Based on our previews work, a compact representation of the patch named patch signature is applied here. A similarity measure of two patches is then derived based on their signatures. To achieve fast patch matching in large protein databases, a match-and-expand strategy is proposed. Given a query patch, a set of small k-sized matching patches, called candidate patches, is generated in match stage. The candidate patches are further filtered by enlarging k in expand stage. Our extensive experimental results demonstrate encouraging performances with respect to this biologically critical but previously computationally prohibitive problem.