Analysis of Compact Features for RGB-D Visual Search

被引:5
作者
Petrelli, Alioscia [1 ]
Pau, Danilo [2 ]
Di Stefano, Luigi [1 ]
机构
[1] Univ Bologna, Bologna, Italy
[2] ST Microelect, Agrate Brianza, Italy
来源
IMAGE ANALYSIS AND PROCESSING - ICIAP 2015, PT II | 2015年 / 9280卷
关键词
RGB-D visual search; Binary hash codes; Deep learning;
D O I
10.1007/978-3-319-23234-8_2
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Anticipating the oncoming integration of depth sensing into mobile devices, we experimentally compare different compact features for representing RGB-D images in mobile visual search. Experiments on 3 state-of-the-art datasets, addressing both category and instance recognition, show how Deep Features provided by Convolutional Neural Networks better represent appearance information, whereas shape is more effectively encoded through Kernel Descriptors. Moreover, our evaluation suggests that learning to weight the relative contribution of depth and appearance is key to deploy effectively depth sensing in forthcoming mobile visual search scenarios.
引用
收藏
页码:14 / 24
页数:11
相关论文
共 22 条
[1]  
[Anonymous], ADV NEURAL INFORM PR
[2]  
[Anonymous], C COMP VIS PATT REC
[3]  
Bo L., 2011, INTELLIGENT ROBOTS S
[4]  
Browatzki B., 2011, INT C COMP VIS WORKS
[5]   BRIEF: Binary Robust Independent Elementary Features [J].
Calonder, Michael ;
Lepetit, Vincent ;
Strecha, Christoph ;
Fua, Pascal .
COMPUTER VISION-ECCV 2010, PT IV, 2010, 6314 :778-792
[6]  
Chandrasekhar V., 2011, INT J COMPUTER VISIO
[7]  
Chandrasekhar V., 2010, INT C PATT REC
[8]  
Chandrasekhar V., 2014, DAT COMPR C
[9]  
Dudani S. A., 1976, IEEE Transactions on Systems, Man and Cybernetics, VSMC-6, P325, DOI 10.1109/TSMC.1976.5408784
[10]   Mobile Visual Search [Linking the virtual and physical worlds] [J].
Girod, Bernd ;
Chandrasekhar, Vijay ;
Chen, David M. ;
Cheung, Ngai-Man ;
Grzeszczuk, Radek ;
Reznik, Yuriy ;
Takacs, Gabriel ;
Tsai, Sam S. ;
Vedantham, Ramakrishna .
IEEE SIGNAL PROCESSING MAGAZINE, 2011, 28 (04) :61-76