Feature hashing

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

Robust Hash Functions for Visual Data: An Experimental Comparison

Abstract: Robust hash functions for visual data need a feature extraction mechanism to rely on. We experimentally compare spatial and transform domain feature extraction techniques and identify the global DCT combined with the cryptographic hash function MD-5 to be suited for visual hashing. This scheme offers robustness against JPEG2000 and JPEG compression and qualitative sensitivity to intentional global and local image alterations.

Pseudo-associative store with hardware hashing

Abstract: In the data-flow model of computation instruction, execution is determined by the availability of data rather than by an explicit or implicit sequential flow of control. One of the major problems in the architectural design of a data-flow computer is the detection of the availability of data. This problem is compounded if the data carry context information as well as pointers to the instructions that will use them; an instruction is then executable when all data directed to it within the same context are present. The solution adopted in the Manchester design is to limit the maximum number of operands of an instruction to two, and to use associative storage techniques to detect the presence of data. The use of true content addressable memory is precluded by its small density and high cost, and therefore a pseudo-associative store using hardware hashing techniques and implemented with conventional random-access memory is employed. The concept of sequence in the data-flow model of computation is unimportant; as a result search operations do not have to be resolved in the same sequence that the store is interrogated. This suggests a design which uses a main parallel hash table and a separate overflow mechanism operating in parallel. In this manner, an overflow search need not halt the progress of further main hash table searches. A pseudo-associative store results whose average access time is very close to the cycle time of the original randon-access memory.

On security threats for robust perceptual hashing

Abstract: Perceptual hashing has to deal with the constraints of robustness, accuracy and security. After modeling the process of hash extraction and the properties involved in this process, two different security threats are studied, namely the disclosure of the secret feature space and the tampering of the hash. Two different approaches for performing robust hashing are presented: Random-Based Hash (RBH) where the security is achieved using a random projection matrix and Content-Based Hash (CBH) were the security relies on the difficulty to tamper the hash. As for digital watermarking, different security setups are also devised: the Batch Hash Attack, the Group Hash Attack, the Unique Hash Attack and the Sensitivity Attack. A theoretical analysis of the information leakage in the context of Random-Based Hash is proposed. Finally, practical attacks are presented: (1) Minor Component Analysis is used to estimate the secret projection of Random-Based Hashes and (2) Salient point tampering is used to tamper the hash of Content-Based Hashes systems.

Accelerated large scale optimization by concomitant hashing

Abstract: Traditional locality-sensitive hashing (LSH) techniques aim to tackle the curse of explosive data scale by guaranteeing that similar samples are projected onto proximal hash buckets. Despite the success of LSH on numerous vision tasks like image retrieval and object matching, however, its potential in large-scale optimization is only realized recently. In this paper we further advance this nascent area. We first identify two common operations known as the computational bottleneck of numerous optimization algorithms in a large-scale setting, i.e., min/max inner product. We propose a hashing scheme for accelerating min/max inner product, which exploits properties of order statistics of statistically correlated random vectors. Compared with other schemes, our algorithm exhibits improved recall at a lower computational cost. The effectiveness and efficiency of the proposed method are corroborated by theoretic analysis and several important applications. Especially, we use the proposed hashing scheme to perform approximate l1 regularized least squares with dictionaries with millions of elements, a scale which is beyond the capability of currently known exact solvers. Nonetheless, it is highlighted that the focus of this paper is not on a new hashing scheme for approximate nearest neighbor problem. It exploits a new application for the hashing techniques and proposes a general framework for accelerating a large variety of optimization procedures in computer vision.