R2FP: Rich and robust feature pooling for mining visual data
Conference Proceeding Article
The human visual system proves smart in extracting both global and local features. Can we design a similar way for unsupervised feature learning? In this paper, we propose a novel pooling method within an unsupervised feature learning framework, named Rich and Robust Feature Pooling (R2FP), to better explore rich and robust representation from sparse feature maps of the input data. Both local and global pooling strategies are further considered to instantiate such a method and intensively studied. The former selects the most conductive features in the sub-region and summarizes the joint distribution of the selected features, while the latter is utilized to extract multiple resolutions of features and fuse the features with a feature balancing kernel for rich representation. Extensive experiments on several image recognition tasks demonstrate the superiority of the proposed techniques.
Autoencoder, Pooling, Representation learning
Computer Sciences | Databases and Information Systems
Data Management and Analytics
ICDM 2015: Proceedings of the 15th IEEE International Conference on Data Mining: Atlantic City, NJ, November 14-17, 2015
City or Country
XIONG, Wei; DU, Bo; ZHANG, Lefei; HU, Ruimin; BIAN, Wei; SHEN, Jialie; and TAO, Dacheng.
R2FP: Rich and robust feature pooling for mining visual data. (2015). ICDM 2015: Proceedings of the 15th IEEE International Conference on Data Mining: Atlantic City, NJ, November 14-17, 2015. 469-478. Research Collection School Of Information Systems.
Available at: http://ink.library.smu.edu.sg/sis_research/3539