Fracking Deep Convolutional Image Descriptors

Abstract

In this paper we propose a novel framework for learning local image descriptors in a discriminative manner. For this purpose we explore a siamese architecture of Deep Convolutional Neural Networks (CNN), with a Hinge embedding loss on the L2 distance between descriptors. Since a siamese architecture uses pairs rather than single image patches to train, there exist a large number of positive samples and an exponential number of negative samples. We propose to explore this space with a stochastic sampling of the training set, in combination with an aggressive mining strategy over both the positive and negative samples which we denote as "fracking". We perform a thorough evaluation of the architecture hyper-parameters, and demonstrate large performance gains compared to both standard CNN learning strategies, hand-crafted image descriptors like SIFT, and the state-of-the-art on learned descriptors: up to 2.5x vs SIFT and 1.5x vs the state-of-the-art in terms of the area under the curve (AUC) of the Precision-Recall curve.

References in corpus (1)

• An Analysis of the Connections Between Layers of Deep Neural Networks

Cited by in corpus (12)

• Recent Advances in Convolutional Neural Networks
• Local Similarity-Aware Deep Feature Embedding
• C-RPNs: Promoting Object Detection in real world via a Cascade Structure of Region Proposal Networks
• Unsupervised Hard Example Mining from Videos for Improved Object Detection
• A-Fast-RCNN: Hard Positive Generation via Adversary for Object Detection
• SampleAhead: Online Classifier-Sampler Communication for Learning from Synthesized Data
• Training Multi-organ Segmentation Networks with Sample Selection by Relaxed Upper Confident Bound
• Better with Less: A Data-Active Perspective on Pre-Training Graph Neural Networks
• Monocular Real-Time Volumetric Performance Capture
• Two-Stream Region Convolutional 3D Network for Temporal Activity Detection
• Improved Hard Example Mining by Discovering Attribute-based Hard Person Identity
• STADB: A Self-Thresholding Attention Guided ADB Network for Person Re-identification