Abstract
In this paper, we propose a content-based motion retrieval (CBMR) algorithm, where many-to-many matching method, weighted graph matching, is employed for comparison between two motions. Our novel points can be described as: (1) A selection approach of representative frames (RF) is presented, in this work, each motion is represented by a set of sequence frames, representative frames are first selected from the motions based on Fuzzy clustering and the corresponding initial weights are provided. (2) The RF-based weighted graph model (RF-WGM) is constructed, and a revised KM (Kuhn---Munkres) algorithm is used to solve maximum matching problem of weighted graph. The RF-WGM matching result is used to measure the similarity between two motions. Experimental results and comparison with existing methods show the effectiveness of the proposed algorithm.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have