Deep Supervised Hashing with Triplet Labels (2016-11-20T00:00:00.000000Z)

TL;DR

A triplet label based deep hashing method which aims to maximize the likelihood of the given triplet labels, which outperforms all the baselines on CIFAR-10 and NUS-WIDE datasets, including the state-of-the-art method DPSH.

Abstract

Hashing is one of the most popular and powerful approximate nearest neighbor search techniques for large-scale image retrieval. Most traditional hashing methods first represent images as off-the-shelf visual features and then produce hashing codes in a separate stage. However, off-the-shelf visual features may not be optimally compatible with the hash code learning procedure, which may result in sub-optimal hash codes. Recently, deep hashing methods have been proposed to simultaneously learn image features and hash codes using deep neural networks and have shown superior performance over traditional hashing methods. Most deep hashing methods are given supervised information in the form of pairwise labels or triplet labels. The current state-of-the-art deep hashing method DPSH [1], which is based on pairwise labels, performs image feature learning and hash code learning simultaneously by maximizing the likelihood of pairwise similarities. Inspired by DPSH [1], we propose a triplet label based deep hashing method which aims to maximize the likelihood of the given triplet labels. Experimental results show that our method outperforms all the baselines on CIFAR-10 and NUS-WIDE datasets, including the state-of-the-art method DPSH [1] and all the previous triplet label based deep hashing methods.

Authors

Kris M. Kitani

4 papers

Xiaofang Wang

1 papers

Yi Shi

1 papers

TL;DR

Abstract

Authors

References36 items

Column Sampling Based Discrete Supervised Hashing

Deep Residual Learning for Image Recognition

Feature Learning Based Deep Supervised Hashing with Pairwise Labels

Bit-Scalable Deep Hashing With Regularized Similarity Learning for Image Retrieval and Person Re-Identification

Scalable Graph Hashing with Feature Transformation

Deep learning of binary hash codes for fast image retrieval

Deep hashing for compact binary codes learning

Simultaneous feature learning and hash coding with deep neural networks

Supervised Discrete Hashing

Deep semantic ranking based hashing for multi-label image retrieval

Discrete Graph Hashing

Supervised hashing with latent factor models

Supervised Hashing for Image Retrieval via Image Representation Learning

Return of the Devil in the Details: Delving Deep into Convolutional Nets

Fast Supervised Hashing with Decision Trees for High-Dimensional Data

Learning Hash Codes with Listwise Supervision

Learning Hash Functions Using Column Generation

Isotropic Hashing

ImageNet classification with deep convolutional neural networks

Supervised hashing with kernels

Minimal Loss Hashing for Compact Binary Codes

Hashing with Graphs

Iterative quantization: A procrustean approach to learning binary codes

Sequential Projection Learning for Hashing with Compact Codes

Learning to Hash with Binary Reconstructive Embeddings

Locality-sensitive binary codes from shift-invariant kernels

NUS-WIDE: a real-world web image database from National University of Singapore

Semantic hashing

Spectral Hashing

Near-Optimal Hashing Algorithms for Approximate Nearest Neighbor in High Dimensions

Distinctive Image Features from Scale-Invariant Keypoints

Modeling the Shape of the Scene: A Holistic Representation of the Spatial Envelope

Learning Multiple Layers of Features from Tiny Images

The above traditional hashing methods using features extracted by CNN-F network [26] pre-trained

Traditional supervised hashing methods using hand-crafted features

Pairwise label based deep hashing methods

Field of Study

Venue Information

Name

Type

URL

Alternate Names