Feature hashing

About: Feature hashing is a research topic. Over the lifetime, 993 publications have been published within this topic receiving 51462 citations.

Hierarchical deep hashing for image retrieval

Abstract: We present a new method to generate efficient multi-level hashing codes for image retrieval based on the deep siamese convolutional neural network (DSCNN). Conventional deep hashing methods trade off the capability of capturing highly complex and nonlinear semantic information of images against very compact hash codes, usually leading to high retrieval efficiency but with deteriorated accuracy. We alleviate the restrictive compactness requirement of hash codes by extending them to a two-level hierarchical coding scheme, in which the first level aims to capture the high-level semantic information extracted by the deep network using a rich encoding strategy, while the subsequent level squeezes them to more global and compact codes. At running time, we adopt an attention-based mechanism to select some of its most essential bits specific to each query image for retrieval instead of using the full hash codes of the first level. The attention-based mechanism is based on the guides of hash codes generated by the second level, taking advantage of both local and global properties of deep features. Experimental results on various popular datasets demonstrate the advantages of the proposed method compared to several state-of-the-art methods.

A multi-label least-squares hashing for scalable image search

Abstract: Recently, hashing methods have attracted more and more attentions for their effectiveness in large scale data search, e.g., images and videos data. etc. For different s-cenarios, unsupervised, supervised and semi-supervised hashing methods have been proposed. Especially, when semantic information is available, supervised hashing methods show better performance than unsupervised ones. In many practical applications, one sample usually has more than one label, which has been considered by multi-label learning. However, few supervised hashing methods consider such scenario. In this paper, we propose a Multi-label Least-Squares Hashing (MLSH) method for multi-label data hashing. It can directly work well on multi-label data; moreover, unlike other hashing methods which directly learn hashing function-s on original data, MLSH first utilizes the equivalen-t form of CCA and Least-Squares to project original multi-label data into lower-dimensional space; then, in the lower-dimensional space, it learns the project matrix and gets final binary codes of data. MLSH is tested on NUS-WIDE and CIFAR-100 which are widely used for searching task. The results show that MLSH outperforms several state-of-the-art hashing methods including supervised and unsupervised methods.

Coordinate discrete optimization for efficient cross-view image retrieval

Abstract: Learning compact hash codes has been a vibrant research topic for large-scale similarity search owing to the low storage cost and expedited search operation. A recent research thrust aims to learn compact codes jointly from multiple sources, referred to as cross-view (or cross-modal) hashing in the literature. The main theme of this paper is to develop a novel formulation and optimization scheme for cross-view hashing. As a key differentiator, our proposed method directly conducts optimization on discrete binary hash codes, rather than relaxed continuous variables as in existing cross-view hashing methods. This way relaxation-induced search accuracy loss can be avoided. We attack the crossview hashing problem by simultaneously capturing semantic neighboring relations and maximizing the generative probability of the learned hash codes in each view. Specifically, to enable effective optimization on discrete hash codes, the optimization proceeds in a block coordinate descent fashion. Each iteration sequentially updates a single bit with others clamped. We transform the resultant sub-problem into an equivalent, more tractable quadratic form and devise an active set based solver on the discrete codes. Rigorous theoretical analysis is provided for the convergence and local optimality condition. Comprehensive evaluations are conducted on three image benchmarks. The clearly superior experimental results faithfully prove the merits of the proposed method.

Abstract Zobrist hashing: an efficient work distribution method for parallel best-first search

Abstract: Hash Distributed A* (HDA*) is an efficient parallel best first algorithm that asynchronously distributes work among the processes using a global hash function. Although Zobrist hashing, the standard hash function used by HDA*, achieves good load balance for many domains, it incurs significant communication overhead since it requires many node transfers among threads. We propose Abstract Zobrist hashing, a new work distribution method for parallel search which reduces node transfers and mitigates communication overhead by using feature projection functions. We evaluate Abstract Zobrist hashing for multicore HDA*, and show that it significantly outperforms previous work distribution methods.

Probabilistic Methods in State Space Analysis

Abstract: Several methods have been developed to validate the correctness and performance of hard- and software systems. One way to do this is to model the system and carry out a state space exploration in order to detect all possible states. In this paper, a survey of existing probabilistic state space exploration methods is given. The paper starts with a thorough review and analysis of bitstate hashing, as introduced by Holzmann. The main idea of this initial approach is the mapping of each state onto a specific bit within an array by employing a hash function. Thus a state is represented by a single bit, rather than by a full descriptor. Bitstate hashing is efficient concerning memory and runtime, but it is hampered by the non deterministic omission of states. The resulting positive probability of producing wrong results is due to the fact that the mapping of full state descriptors onto much smaller representatives is not injective. – The rest of the paper is devoted to the presentation, analysis, and comparison of improvements of bitstate hashing, which were introduced in order to lower the probability of producing a wrong result, but maintaining the memory and runtime efficiency. These improvements can be mainly grouped into two categories: The approaches of the first group, the so called multiple hashing schemes, employ multiple hash functions on either a single or on multiple arrays. The approaches of the remaining category follow the idea of hash compaction. I.e. the diverse schemes of this category store a hash value for each detected state, rather than associating a single or multiple bit positions with it, leading to persuasive reductions of the probability of error if compared to the original bitstate hashing scheme.