DocumentCode :
2115465
Title :
Scalable classifiers for Internet vision tasks
Author :
Yeh, Tom ; Lee, John J. ; Darrell, Trevor
Author_Institution :
EECS & CSAIL, Massachesettes Inst. of Technol., Cambridge, MA
fYear :
2008
fDate :
23-28 June 2008
Firstpage :
1
Lastpage :
8
Abstract :
Object recognition systems designed for Internet applications typically need to adapt to userspsila needs in a flexible fashion and scale up to very large data sets. In this paper, we analyze the complexity of several multiclass SVM-based algorithms and highlight the computational bottleneck they suffer at test time: comparing the input image to every training image. We propose an algorithm that overcomes this bottleneck; it offers not only the efficiency of a simple nearest-neighbor classifier, by voting on class labels based on the k nearest neighbors quickly determined by a vocabulary tree, but also the recognition accuracy comparable to that of a complex SVM classifier, by incorporating SVM parameters into the voting scores incrementally accumulated from individual image features. Empirical results demonstrate that adjusting votes by relevant support vector weights can improve the recognition accuracy of a nearest-neighbor classifier without sacrificing speed. Compared to existing methods, our algorithm achieves a ten-fold speed increase while incurring an acceptable accuracy loss that can be easily offset by showing about two more labels in the result. The speed, scalability, and adaptability of our algorithm makes it suitable for Internet vision applications.
Keywords :
Internet; computational complexity; computer vision; image classification; image retrieval; learning (artificial intelligence); object recognition; support vector machines; vocabulary; Internet vision tasks; image retrieval; k nearest neighbors classifier; multiclass SVM-based algorithm complexity; object recognition systems; scalable classifiers; training image; vocabulary tree; Algorithm design and analysis; Classification tree analysis; Image analysis; Internet; Nearest neighbor searches; Object recognition; Support vector machine classification; Support vector machines; Testing; Voting;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Computer Vision and Pattern Recognition Workshops, 2008. CVPRW '08. IEEE Computer Society Conference on
Conference_Location :
Anchorage, AK
ISSN :
2160-7508
Print_ISBN :
978-1-4244-2339-2
Electronic_ISBN :
2160-7508
Type :
conf
DOI :
10.1109/CVPRW.2008.4562958
Filename :
4562958
Link To Document :
بازگشت