Iterative keyframe selection by orthogonal subspace projection

Abstract

Recent developments on sparse dictionary selection have demonstrated promising results for Video Summarization (VS). However, the convex relaxation based solution cannot ensure the sparsity of the dictionary directly. In this paper, a selection matrix is proposed to model the VS problem, according to which the L 0 norm of this selection matrix is imposed to ensure sparsity directly. As a result, a computational efficient Orthogonal Subspace Projection (OSP) based Iterative Keyframe Selection (IKS) algorithm is proposed for VS. In addition, a Percentage Of Reconstruction (POR) criterion is proposed to provide an intuitive and flexible control of the length of final video summaries even without prior knowledge of a given video. Experimental results on a popular benchmark dataset demonstrate that our proposed algorithm outperforms the state-of-the-art methods.

Resource restricted on-line Video Summarization with Minimum Sparse Reconstruction

“…The combination of global and local features for video frame representation has also been utilized for keyframe extraction [12]. The minimum sparse reconstruction based algorithms select a set of keyframes (namely a dictionary) that can reconstruct all the frames of a video [13], [14], [15], [16]. Most of the existing VS algorithms are performed in an off-line manner, and only a minority of them encounter the on-line situation [16], [17], [18], though the on-line methods are important for many applications.…”