sketch-based human motion retrieval via selected 2d geometric posture descriptor论文.pdfVIP

Signal Processing 113 (2015) 1–8 Contents lists available at ScienceDirect Signal Processing journal homepage: /locate/sigpro Sketch-based human motion retrieval via selected 2D geometric posture descriptor Jun Xiao a, Zhangpeng Tang a,n, Yinfu Feng a, Zhidong Xiao b a School of Computer Science, Zhejiang University, Hangzhou 310027, PR China b Bournemouth University, Poole Dorset BH12, United Kingdom a r t i c l e i n f o a b s t r a c t Article history: Sketch-based human motion retrieval is a hot topic in computer animation in recent years. Received 7 September 2014 In this paper, we present a novel sketch-based human motion retrieval method via Received in revised form selected 2-dimensional (2D) Geometric Posture Descriptor (2GPD). Specially, we firstly 5 December 2014 propose a rich 2D pose feature call 2D Geometric Posture Descriptor (2GPD), which is Accepted 5 January 2015 effective in encoding the 2D posture similarity by exploiting the geometric relationships Available online 12 January 2015 among different human body parts. Since the original 2GPD is of high dimension and Keywords: redundant, a semi-supervised feature selection algorithm derived from Laplacian Score is Motion retrieval then adopted to select the most discriminative feature component of 2GPD as feature Sketch-base