Parameter learning for improving binary descriptor matching

Binary descriptors allow fast detection and matching algorithms in computer vision problems. Though binary descriptors can be computed at almost two orders of magnitude faster than traditional gradient based descriptors, they suffer from poor matching accuracy in challenging conditions. In this paper we propose three improvements for binary descriptors in their computation and matching that enhance their performance in comparison to traditional binary and non-binary descriptors without compromising their speed. This is achieved by learning some weights and threshold parameters that allow customized matching under some variations such as lighting and viewpoint. Our suggested improvements can be easily applied to any binary descriptor. We demonstrate our approach on the ORB (Oriented FAST and Rotated BRIEF) descriptor and compare its performance with the traditional ORB and SIFT descriptors on a wide variety of datasets. In all instances, our enhancements outperform standard ORB and are comparable to SIFT.

Authors

• Bharath Sankaran
• Srikumar Ramalingam
• Yuichi Taguchi

Keywords

• Hamming distance
• Lighting
• Standards
• Training data
• Feature extraction
• Measurement
• Real-time systems
• Binary Descriptors
• Descriptor Matching
• Matching Accuracy
• Training Set
• Convolutional Neural Network
• Computational Efficiency
• Image Registration
• Scale Variation
• Optimal Threshold
• Target Image
• Exhaustive Search
• Computational Speed
• Source Images
• Binary Code
• Learned Weights
• Variable Light
• Sum Of Squared Differences
• Correct Matches
• KITTI Dataset
• Stereo Pairs
• Calculation Of Descriptors
• Feature Matching
• Image Patches
• Lookup Table
• Object Detection
• Image Distortion
• Integer Values