Receptive Fields Selection for Binary Feature Description
- 16 April 2014
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Image Processing
- Vol. 23 (6), 2583-2595
- https://doi.org/10.1109/tip.2014.2317981
Abstract
Feature description for local image patch is widely used in computer vision. While the conventional way to design local descriptor is based on expert experience and knowledge, learning-based methods for designing local descriptor become more and more popular because of their good performance and data-driven property. This paper proposes a novel data-driven method for designing binary feature descriptor, which we call receptive fields descriptor (RFD). Technically, RFD is constructed by thresholding responses of a set of receptive fields, which are selected from a large number of candidates according to their distinctiveness and correlations in a greedy way. Using two different kinds of receptive fields (namely rectangular pooling area and Gaussian pooling area) for selection, we obtain two binary descriptors RFD R and RFD G accordingly. Image matching experiments on the well-known patch data set and Oxford data set demonstrate that RFD significantly outperforms the state-of-the-art binary descriptors, and is comparable with the best float-valued descriptors at a fraction of processing time. Finally, experiments on object recognition tasks confirm that both RFD R and RFD G successfully bridge the performance gap between binary descriptors and their floating-point competitors.Keywords
Funding Information
- National Natural Science Foundation of China (61203277, 61272394, 91338202)
- Beijing Natural Science Foundation (4142057)
- Tsinghua National Laboratory for Information Science and Technology Cross-Discipline Foundation
This publication has 33 references indexed in Scilit:
- RLBP: Robust Local Binary PatternPublished by British Machine Vision Association and Society for Pattern Recognition ,2013
- Rotationally Invariant Descriptors Using Intensity Order PoolingIEEE Transactions on Pattern Analysis and Machine Intelligence, 2011
- Discriminative Learning of Local Image DescriptorsIEEE Transactions on Pattern Analysis and Machine Intelligence, 2010
- Robust order-based methods for feature descriptionPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2010
- Kernelized locality-sensitive hashing for scalable image searchPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2009
- Semantic hashingInternational Journal of Approximate Reasoning, 2009
- Description of interest regions with local binary patternsPattern Recognition, 2009
- Dynamic Texture Recognition Using Local Binary Patterns with an Application to Facial ExpressionsIEEE Transactions on Pattern Analysis and Machine Intelligence, 2007
- A performance evaluation of local descriptorsIEEE Transactions on Pattern Analysis and Machine Intelligence, 2005
- A comparative study of texture measures with classification based on featured distributionsPattern Recognition, 1996