Search

Thanks to the efficient retrieval speed and low storage consumption, learning to hash has been widely used in visual retrieval tasks. However, the known hashing methods assume that the query and retrieval samples lie in homogeneous feature space within the same domain. As a result, they cannot be directly applied to heterogeneous cross-domain retrieval. In this article, we propose a generalized image transfer retrieval (GITR) problem, which encounters two crucial bottlenecks: 1) the query and retrieval samples may come from different domains, leading to an inevitable domain distribution gap and 2) the features of the two domains may be heterogeneous or misaligned, bringing up an additional feature gap. To address the GITR problem, we propose an asymmetric transfer hashing (ATH) framework with its unsupervised/semisupervised/supervised realizations. Specifically, ATH characterizes the domain distribution gap by the discrepancy between two asymmetric hash functions, and minimizes the feature gap with the help of a novel adaptive bipartite graph constructed on cross-domain data. By jointly optimizing asymmetric hash functions and the bipartite graph, not only can knowledge transfer be achieved but information loss caused by feature alignment can also be avoided. Meanwhile, to alleviate negative transfer, the intrinsic geometrical structure of single-domain data is preserved by involving a domain affinity graph. Extensive experiments on both single-domain and cross-domain benchmarks under different GITR subtasks indicate the superiority of our ATH method in comparison with the state-of-the-art hashing methods.

• We propose a novel deep supervised hashing framework called DSAH, which adopts an interesting similarity preserving strategy based on distance-similarity product function. • DSAH subtly incorporates class prior of data into the quantization stage, thereby providing discriminative feedback for network training. • A dual semantic regression loss supported by a simple yet effective label mechanism is designed to characterize the intra-class compactness and inter-class separability of data. Recently, deep hashing methods have been widely used in image retrieval task. Most existing deep hashing approaches adopt one-to-one quantization to reduce information loss. However, such class-unrelated quantization cannot give discriminative feedback for network training. In addition, these methods only utilize single label to integrate supervision information of data for hashing function learning, which may result in inferior network generalization performance and relatively low-quality hash codes since the inter-class information of data is totally ignored. In this paper, we propose a dual semantic asymmetric hashing (DSAH) method, which generates discriminative hash codes under threefold constraints. Firstly, DSAH utilizes class prior to conduct class structure quantization so as to transmit class information during the quantization process. Secondly, a simple yet effective label mechanism is designed to characterize both the intra-class compactness and inter-class separability of data, thereby achieving semantic-sensitive binary code learning. Finally, a meaningful pairwise similarity preserving loss is devised to minimize the distances between class-related network outputs based on an affinity graph. With these three main components, high-quality hash codes can be generated through network. Extensive experiments conducted on various data sets demonstrate the superiority of DSAH in comparison with state-of-the-art deep hashing methods. [ABSTRACT FROM AUTHOR]

Sparse discriminative projection learning has attracted much attention due to its good performance in recognition tasks. In this article, a framework called generalized embedding regression (GER) is proposed, which can simultaneously perform low-dimensional embedding and sparse projection learning in a joint objective function with a generalized orthogonal constraint. Moreover, the label information is integrated into the model to preserve the global structure of data, and a rank constraint is imposed on the regression matrix to explore the underlying correlation structure of classes. Theoretical analysis shows that GER can obtain the same or approximate solution as some related methods with special settings. By utilizing this framework as a general platform, we design a novel supervised feature extraction approach called jointly sparse embedding regression (JSER). In JSER, we construct an intrinsic graph to characterize the intraclass similarity and a penalty graph to indicate the interclass separability. Then, the penalty graph Laplacian is used as the constraint matrix in the generalized orthogonal constraint to deal with interclass marginal points. Moreover, the $L_{2,1}$ -norm is imposed on the regression terms for robustness to outliers and data’s variations and the regularization term for jointly sparse projection learning, leading to interesting semantic interpretability. An effective iterative algorithm is elaborately designed to solve the optimization problem of JSER. Theoretically, we prove that the subproblem of JSER is essentially an unbalanced Procrustes problem and can be solved iteratively. The convergence of the designed algorithm is also proved. Experimental results on six well-known data sets indicate the competitive performance and latent properties of JSER. [ABSTRACT FROM AUTHOR]

ABSTRACTThe influential factors of the generation of kitchen solid waste (KSW) are vital to design policies that target effective KSW reductions and achieve sustainable KSW management, but has not been extensively studied. The study aims at the first step of KSW management only, i.e., generation, which is the first KSW analysis survey in Shanghai to comprehensively investigate the characteristics and influential factors of KSW generation (KSWG). The data were collected from August to October 2019, which resulted in a total of 433 usable questionnaires regarding KSWG in Shanghai, China. The average rate of KSWG is 6.038 kg/household/week or 0.265 kg/person/day. The average per-person number of KSWG bags is 0.295. The regression results show that the characteristics of KSWG are related to household size, household income, the habit of ordering takeout, district of residence, and meticulous consumption culture. Some interesting findings are observed in the present study. Firstly, there is a nonlinear relationship between KSWG and household size, which can be explained by “common consumption”. Secondly, there is a nonlinear relationship between KSWG and household income, which can be explained by the Engel coefficient, which also provides supporting evidence for the environmental Kuznets curve (EKC) hypothesis. Finally, the habit of ordering takeout, district of residence, and meticulous consumption culture are all influential factors of KSWG. The results of the study can improve public awareness and knowledge of KSWG, and suggestions on the minimization of KSWG are put forward, thus will serve as a new reference for the sustainable KSW management of other cities, especially cities in developing countries.Implications: In the practice of sustainable waste management, KSWG is a complex issue that requires a broad analytic approach that considers several factors simultaneously. From a practical perspective, policy decision-makers should pay attention to at least four important actions to reduce KSWG: Firstly, compared with urban areas, rural areas are important areas for reducing KSWG, which should be given sufficient attention. Secondly, compared with low-income households, high-income households are the focus of KSWG reduction and should be considered. Thirdly, meticulous consumption culture plays an important role in promoting the reduction of KSWG, and this culture should be promoted and cultivated in the formulation of actual waste management policies. This is also an important way to reduce KSWG and should be fully considered. Finally, in the pilot process of domestic waste separation in Chinese cities, 3 R (Reduce, Reuse, and Recycle) waste management concept and method in circular economy can be applied to KSW treatment, and adequate financial support is necessary. Through strengthening the management of KSW, the potential value of KSW can be further developed, so as to realize KSWG reduction and resource utilization.

• A general regression framework called RLRR is proposed. • A meaningful dynamic graph with probabilistic weights is adopted in RLRR. • The targets are redirected to a variable matrix for the sake of margin amplication. • Two novel regression methods called RLRRM and RLRRP are developed based on RLRR. • Two ecient iterative algorithms are designed to optimize the objective functions. Least squares regression (LSR) has attracted widespread attention in the fields of statistics, machine learning, and pattern recognition. However, it utilizes strict zero-one regression targets, which leads to inferior performance on classification tasks. Furthermore, LSR ignores the local manifold structures of data and lacks robustness. To address these issues, this paper proposes a general regression framework called RLRR, where a low-rank constraint is imposed on regression matrices to explore the underlying correlation structures of classes. Strict zero-one regression targets are redirected to more feasible variable matrices for the purpose of margin amplification of different classes. Additionally, rather than using a pre-constructed weighted graph, the proposed framework dynamically updates the neighborhood structures of data to preserve original manifold structures. By utilizing this framework as a general platform, we developed two dynamic neighborhood-structure-based regression models called RLRRM and RLRRP. RLRRM integrates a reconstruction error minimization term into the proposed RLRR framework, whereas RLRRP aims to preserve the local geometric structures of data in a low-dimensional subspace. Both RLRRM and RLRRP use the ℓ 2 , 1 -norm penalty to replace the traditional F -norm penalty for the projection matrix for the sake of self-adaptive feature selection. Instead of directly solving the resultant optimization problems with non-convex constraints, we adopt the variable-splitting and penalty techniques to derive an equivalent solution. Analysis of the corresponding convergence and computational complexity characteristics is also presented. Extensive experiments on several well-known datasets demonstrate the promising performance of the proposed models. [ABSTRACT FROM AUTHOR]