Motion History Volumes for Free Viewpoint Action Recognition
作者： Rémi Ronfard , Daniel Weinland , Edmond Boyer
DOI:
关键词: Viewpoints 、 Computer science 、 Action (philosophy) 、 Artificial intelligence 、 Motion history 、 Variety (cybernetics) 、 Interpretation (logic) 、 Computer vision 、 Representation (systemics) 、 Gesture 、 Cinematography
摘要: Action recognition is an important and challenging topic in computer vision, with many applications including video surveillance, automated cinematography understanding of social interaction. Yet, most current work gesture or action interpretation remains rooted viewdependent representations. This paper introduces Motion History Volumes (MHV) as a free-viewpoint representation for human actions the case multiple calibrated, background-subtracted, cameras. We present algorithms computing, aligning comparing MHVs different performed by people variety viewpoints. Alignment comparisons are efficiently using Fourier transforms cylindrical coordinates around vertical axis. Preliminary results indicate that this can be used to learn recognize basic classes, independently gender, body size viewpoint.
参考文章(23)
Michael Kazhdan, Szymon Rusinkiewicz, Thomas Funkhouser, Rotation invariant spherical harmonic representation of 3D shape descriptors symposium on geometry processing. pp. 156- 164 ,(2003) , 10.5555/882370.882392
Atsuhiro Kojima, Takeshi Tamura, Kunio Fukunaga, Natural Language Description of Human Activities from Video Images Based on Concept Hierarchy of Actions International Journal of Computer Vision. ,vol. 50, pp. 171- 184 ,(2002) , 10.1023/A:1020346032608
Daniel Heesch, Stefan Rüger, Combining Features for Content-Based Sketch Retrieval - A Comparative Evaluation of Retrieval Performance european conference on information retrieval. pp. 41- 52 ,(2002) , 10.1007/3-540-45886-7_3
Yaser S. Abu-Mostafa, Demetri Psaltis, Image Normalization by Complex Moments IEEE Transactions on Pattern Analysis and Machine Intelligence. ,vol. PAMI-7, pp. 46- 55 ,(1985) , 10.1109/TPAMI.1985.4767617
A.E Grace, M Spann, A comparison between Fourier-Mellon descriptors and moment based features for invariant object recognition using neural networks Pattern Recognition Letters. ,vol. 12, pp. 635- 643 ,(1991) , 10.1016/0167-8655(91)90018-H
Representation and recognition of the movements of shapes. Proceedings of The Royal Society B: Biological Sciences. ,vol. 214, pp. 501- 524 ,(1982) , 10.1098/RSPB.1982.0024
Ray Jackendoff, On beyond Zebra: The relation of linguistic and visual information ☆ Cognition. ,vol. 26, pp. 89- 114 ,(1987) , 10.1016/0010-0277(87)90026-6
T. Syeda-Mahmood, A. Vasilescu, S. Sethi, Recognizing action events from multiple viewpoints Proceedings IEEE Workshop on Detection and Recognition of Events in Video. pp. 64- 72 ,(2001) , 10.1109/EVENT.2001.938868
Firooz A. Sadjadi, Ernest L. Hall, Three-Dimensional Moment Invariants IEEE Transactions on Pattern Analysis and Machine Intelligence. ,vol. 2, pp. 127- 136 ,(1980) , 10.1109/TPAMI.1980.4766990
A. Laurentini, The visual hull concept for silhouette-based image understanding IEEE Transactions on Pattern Analysis and Machine Intelligence. ,vol. 16, pp. 150- 162 ,(1994) , 10.1109/34.273735