Random Binary Mappings for Kernel Learning and Efficient SVM

TL;DR

This paper presents a novel randomized binary mapping kernel that enhances SVM efficiency and adaptability across various image descriptors, reducing kernel selection complexity in computer vision tasks.

Contribution

Introduces a new kernel learned from randomized binary mappings, improving SVM efficiency and flexibility without extensive kernel tuning.

Findings

Achieves performance comparable to specialized kernels for different descriptors.

Reduces computational and memory costs in SVM training.

Demonstrates effectiveness on 6 standard vision benchmarks.

Abstract

Support Vector Machines (SVMs) are powerful learners that have led to state-of-the-art results in various computer vision problems. SVMs suffer from various drawbacks in terms of selecting the right kernel, which depends on the image descriptors, as well as computational and memory efficiency. This paper introduces a novel kernel, which serves such issues well. The kernel is learned by exploiting a large amount of low-complex, randomized binary mappings of the input feature. This leads to an efficient SVM, while also alleviating the task of kernel selection. We demonstrate the capabilities of our kernel on 6 standard vision benchmarks, in which we combine several common image descriptors, namely histograms (Flowers17 and Daimler), attribute-like descriptors (UCI, OSR, and a-VOC08), and Sparse Quantization (ImageNet). Results show that our kernel learning adapts well to the different descriptors types, achieving the performance of the kernels specifically tuned for each image descriptor, and with similar evaluation cost as efficient SVM methods.