Pehlivan, SelenDuygulu, Pınar2016-02-082016-02-082009-09-10http://hdl.handle.net/11693/28664Date of Conference: 27 Sept.-4 Oct. 2009Conference name: IEEE 12th International Conference on Computer Vision Workshops, ICCV Workshops 2009In this study, we present a representation based on a new 3D search technique for volumetric human poses which is then used to recognize actions in three dimensional video sequences. We generate a set of cylinder like 3D kernels in various sizes and orientations. These kernels are searched over 3D volumes to find high response regions. The distribution of these responses are then used to represent a 3D pose. We use the proposed representation for (i) pose retrieval using Nearest Neighbor (NN) based classification and Support Vector Machine (SVM) based classification methods, and for (ii) action recognition on a set of actions using Dynamic Time Warping (DTW) and Hidden Markov Model (HMM) based classification methods. Evaluations on IXMAS dataset supports the effectiveness of such a robust pose representation. ©2009 IEEE.EnglishAction recognitionClassification methodsData setsDynamic time warpingHigh responseHuman poseNearest neighborsSearch techniqueVideo sequencesComputer visionCylinders (shapes)Hidden Markov modelsVideo recordingThree dimensionalConference Paper10.1109/ICCVW.2009.5457722