Two-Stream Deep Hashing With Class-Specific Centers for Supervised Image Search.

08:00 EDT 11th September 2019 | BioPortfolio

Summary of "Two-Stream Deep Hashing With Class-Specific Centers for Supervised Image Search."

Hashing has been widely used for large-scale approximate nearest neighbor search due to its storage and search efficiency. Recent supervised hashing research has shown that deep learning-based methods can significantly outperform nondeep methods. Most existing supervised deep hashing methods exploit supervisory signals to generate similar and dissimilar image pairs for training. However, natural images can have large intraclass and small interclass variations, which may degrade the accuracy of hash codes. To address this problem, we propose a novel two-stream ConvNet architecture, which learns hash codes with class-specific representation centers. Our basic idea is that if we can learn a unified binary representation for each class as a center and encourage hash codes of images to be close to the corresponding centers, the intraclass variation will be greatly reduced. Accordingly, we design a neural network that leverages label information and outputs a unified binary representation for each class. Moreover, we also design an image network to learn hash codes from images and force these hash codes to be close to the corresponding class-specific centers. These two neural networks are then seamlessly incorporated to create a unified, end-to-end trainable framework. Extensive experiments on three popular benchmarks corroborate that our proposed method outperforms current state-of-the-art methods.


Journal Details

This article was published in the following journal.

Name: IEEE transactions on neural networks and learning systems
ISSN: 2162-2388


DeepDyve research library

PubMed Articles [13085 Associated PubMed Articles listed on BioPortfolio]

Deep Class-Wise Hashing: Semantics-Preserving Hashing via Class-Wise Loss.

Deep supervised hashing has emerged as an effective solution to large-scale semantic image retrieval problems in computer vision. Convolutional neural network-based hashing methods typically seek pair...

Interval Type-2 Fuzzy Logic for Semisupervised Multimodal Hashing.

Retrieving nearest neighbors across correlated data in multiple modalities, such as image-text pairs on Facebook and video-tag pairs on YouTube, has become a challenging task due to the huge amount of...

Hashing with Mutual Information.

Binary vector embeddings enable fast nearest neighbor retrieval in large databases of high-dimensional objects, and play an important role in many practical applications, such as image and video retri...

Simultaneous Feature Aggregating and Hashing for Compact Binary Code Learning.

Representing images by compact hash codes is an attractive approach for large-scale content-based image retrieval. In most state-of-the-art hashing-based image retrieval systems, for each image, local...

Siamese Dilated Inception Hashing With Intra-Group Correlation Enhancement for Image Retrieval.

For large-scale image retrieval, hashing has been extensively explored in approximate nearest neighbor search methods due to its low storage and high computational efficiency. With the development of ...

Clinical Trials [5862 Associated Clinical Trials listed on BioPortfolio]

Efficacy of a Supervised Physical Therapy Rehabilitation Program vs Nonsupervised in Non-specific Low Back Pain

This study compares the efficacy of a supervised rehabilitation program vs a non supervised rehabilitation program on pain, disability, quality of life and kinesiophobia in patients with n...

Effect of Deep Brain Stimulation on Lower Urinary Tract Function

The precise mechanisms underlying cerebral regulation of lower urinary tract (LUT) function are still poorly understood. Patients with deep brain stimulation (DBS) offer the unique opportu...

A Comparison of the Stroke Rehabilitation Assessment of Movement, the Functional Independence Measure and the Stroke Impact Scale

The Stroke Rehabilitation Assessment of Movement (STREAM) is a relatively new measurement tool, which measures specific movement deficits (impairments) caused by a stroke. The STREAM is u...

Deep Neuromuscular Relaxation Optimizing Low-Pressure Bariatric Surgery

This study aims to evaluate the use of deep neuromuscular blockade and low-pressure pneumoperitoneum in bariatric surgery. The investigators propose to conduct a double-blind randomized co...

Faith and Community in Action: Increasing Awareness and Management of Depression in African American Communities

Specific Aims/Study Objectives: 1. Evaluate acceptability and implementation of the "Oh Happy Day: Still I Rise Class," which comprise a one-time 4-hour class in which participants will l...

Medical and Biotech [MESH] Definitions

Education centers authorized by the Comprehensive Health Manpower Training Act, 1971, for the training of health personnel in areas where health needs are the greatest. May be used for centers other than those established by the United States act.

DEEP VEIN THROMBOSIS of an upper extremity vein (e.g., AXILLARY VEIN; SUBCLAVIAN VEIN; and JUGULAR VEINS). It is associated with mechanical factors (Upper Extremity Deep Vein Thrombosis, Primary) secondary to other anatomic factors (Upper Extremity Deep Vein Thrombosis, Secondary). Symptoms may include sudden onset of pain, warmth, redness, blueness, and swelling in the arm.

A class of G-protein-coupled receptors that are specific for CANNABINOIDS such as those derived from CANNABIS. They also bind a structurally distinct class of endogenous factors referred to as ENDOCANNABINOIDS. The receptor class may play a role in modulating the release of signaling molecules such as NEUROTRANSMITTERS and CYTOKINES.

A medical facility which provides a high degree of subspecialty expertise for patients from centers where they received SECONDARY CARE.

Health facilities providing therapy and/or rehabilitation for substance-dependent individuals. Methadone distribution centers are included.

Quick Search

DeepDyve research library

Relevant Topic

Antiretroviral therapy
Standard antiretroviral therapy (ART) consists of the combination of at least three antiretroviral (ARV) drugs to maximally suppress the HIV virus and stop the progression of HIV disease. Huge reductions have been seen in rates of death and suffering whe...

Searches Linking to this Article