Circulant Temporal Encoding for Video Retrieval and Temporal Alignment

作者: Matthijs Douze , Jérôme Revaud , Jakob Verbeek , Hervé Jégou , Cordelia Schmid

DOI: 10.1007/S11263-015-0875-0

关键词:

摘要: We address the problem of specific video event retrieval. Given a query event, e.g., concert Madonna, goal is to retrieve other videos same that temporally overlap with query. Our approach encodes frame descriptors jointly represent their appearance and temporal order. It exploits properties circulant matrices efficiently compare in frequency domain. This offers significant gain complexity accurately localizes matching parts videos. The can be compressed domain product quantizer adapted complex numbers. In this case, retrieval performed without decompressing descriptors. also consider alignment set exploit confidence an estimate offset computed for all pairs by our approach. robust algorithm aligns on global timeline maximizing consistent matches. enables synchronous playback given scene.

参考文章(43)
Amir Roshan Zamir, Khurram Soomro, Mubarak Shah, UCF101: A Dataset of 101 Human Actions Classes From Videos in The Wild arXiv: Computer Vision and Pattern Recognition. ,(2012)
Wen-Sheng Chu, Feng Zhou, Fernando De la Torre, Unsupervised temporal commonality discovery european conference on computer vision. pp. 373- 387 ,(2012) , 10.1007/978-3-642-33765-9_27
Hanqing Jiang, Haomin Liu, Ping Tan, Guofeng Zhang, Hujun Bao, 3D Reconstruction of Dynamic Scenes with Multiple Handheld Cameras Computer Vision – ECCV 2012. pp. 601- 615 ,(2012) , 10.1007/978-3-642-33709-3_43
João F. Henriques, Rui Caseiro, Pedro Martins, Jorge Batista, Exploiting the circulant structure of tracking-by-detection with kernels european conference on computer vision. pp. 702- 715 ,(2012) , 10.1007/978-3-642-33765-9_50
Abhijit Mahalanobis, Richard D. Juday, B. V. K. Vijaya Kumar, Correlation Pattern Recognition ,(2005)
Fernando De la Torre, Minh Hoai Nguyen, Maximum Margin Temporal Clustering international conference on artificial intelligence and statistics. pp. 520- 528 ,(2012)
Herve Jegou, Matthijs Douze, Cordelia Schmid, Hamming Embedding and Weak Geometric Consistency for Large Scale Image Search european conference on computer vision. ,vol. 5302, pp. 304- 317 ,(2008) , 10.1007/978-3-540-88682-2_24
Charles Dubout, François Fleuret, Exact Acceleration of Linear Object Detectors Computer Vision – ECCV 2012. pp. 301- 311 ,(2012) , 10.1007/978-3-642-33712-3_22
Matthijs Douze, Hervé Jégou, Cordelia Schmid, Patrick Pérez, Compact video description for copy detection with precise temporal alignment european conference on computer vision. ,vol. 6311, pp. 522- 535 ,(2010) , 10.1007/978-3-642-15549-9_38