SUHA

Abstract: We demonstrate an efficient data-parallel algorithm for building large hash tables of millions of elements in real-time. We consider two parallel algorithms for the construction: a classical sparse perfect hashing approach, and cuckoo hashing, which packs elements densely by allowing an element to be stored in one of multiple possible locations. Our construction is a hybrid approach that uses both algorithms. We measure the construction time, access time, and memory usage of our implementations and demonstrate real-time performance on large datasets: for 5 million key-value pairs, we construct a hash table in 35.7 ms using 1.42 times as much memory as the input data itself, and we can access all the elements in that hash table in 15.3 ms. For comparison, sorting the same data requires 36.6 ms, but accessing all the elements via binary search requires 79.5 ms. Furthermore, we show how our hashing methods can be applied to two graphics applications: 3D surface intersection for moving data and geometric hashing for image matching.

Abstract: Recently, deep neural networks based hashing methods have greatly improved the multimedia retrieval performance by simultaneously learning feature representations and binary hash functions. Inspired by the latest advance in the asymmetric hashing scheme, in this work, we propose a novel Deep Asymmetric Pairwise Hashing approach (DAPH) for supervised hashing. The core idea is that two deep convolutional models are jointly trained such that their output codes for a pair of images can well reveal the similarity indicated by their semantic labels. A pairwise loss is elaborately designed to preserve the pairwise similarities between images as well as incorporating the independence and balance hash code learning criteria. By taking advantage of the flexibility of asymmetric hash functions, we devise an efficient alternating algorithm to optimize the asymmetric deep hash functions and high-quality binary code jointly. Experiments on three image benchmarks show that DAPH achieves the state-of-the-art performance on large-scale image retrieval.

Abstract: Hashing techniques have been intensively investigated for large scale vision applications. Recent research has shown that leveraging supervised information can lead to high quality hashing. However, most existing supervised hashing methods only construct similarity-preserving hash codes. Observing that semantic structures carry complementary information, we propose the idea of cotraining for hashing, by jointly learning projections from image representations to hash codes and classification. Specifically, a novel deep semantic-preserving and ranking-based hashing (DSRH) architecture is presented, which consists of three components: a deep CNN for learning image representations, a hash stream of a binary mapping layer by evenly dividing the learnt representations into multiple bags and encoding each bag into one hash bit, and a classification stream. Mean-while, our model is learnt under two constraints at the top loss layer of hash stream: a triplet ranking loss and orthogonality constraint. The former aims to preserve the relative similarity ordering in the triplets, while the latter makes different hash bit as independent as possible. We have conducted experiments on CIFAR-10 and NUS-WIDE image benchmarks, demonstrating that our approach can provide superior image search accuracy than other state-of-the-art hashing techniques.

Abstract: In this paper, we study the effective semi-supervised hashing method under the framework of regularized learning-based hashing. A nonlinear hash function is introduced to capture the underlying relationship among data points. Thus, the dimensionality of the matrix for computation is not only independent from the dimensionality of the original data space but also much smaller than the one using linear hash function. To effectively deal with the error accumulated during converting the real-value embeddings into the binary code after relaxation, we propose a semi-supervised nonlinear hashing algorithm using bootstrap sequential projection learning which effectively corrects the errors by taking into account of all the previous learned bits holistically without incurring the extra computational overhead. Experimental results on the six benchmark data sets demonstrate that the presented method outperforms the state-of-the-art hashing algorithms at a large margin.