DocumentCode :
253645
Title :
Anytime Recognition of Objects and Scenes
Author :
Karayev, Sergey ; Fritz, Matt ; Darrell, Trevor
fYear :
2014
fDate :
23-28 June 2014
Firstpage :
572
Lastpage :
579
Abstract :
Humans are capable of perceiving a scene at a glance, and obtain deeper understanding with additional time. Similarly, visual recognition deployments should be robust to varying computational budgets. Such situations require Anytime recognition ability, which is rarely considered in computer vision research. We present a method for learning dynamic policies to optimize Anytime performance in visual architectures. Our model sequentially orders feature computation and performs subsequent classification. Crucially, decisions are made at test time and depend on observed data and intermediate results. We show the applicability of this system to standard problems in scene and object recognition. On suitable datasets, we can incorporate a semantic back-off strategy that gives maximally specific predictions for a desired level of accuracy, this provides a new view on the time course of human visual perception.
Keywords :
computer vision; feature extraction; image classification; object recognition; visual perception; anytime recognition ability; computer vision research; feature computation; human visual perception; object recognition; scene recognition; visual architectures; visual recognition deployments; Computer vision; Feature extraction; Hafnium; Logistics; Training; Vectors; Visualization; anytime; budgeted classification; visual recognition;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Computer Vision and Pattern Recognition (CVPR), 2014 IEEE Conference on
Conference_Location :
Columbus, OH
Type :
conf
DOI :
10.1109/CVPR.2014.80
Filename :
6909474
Link To Document :
بازگشت