In recent years, hashing has attracted more and more attention owing to its superior capacity of low storage cost and high query efficiency in large-scale cross-modal retrieval. Benefiting from deep leaning, continuously compelling results in cross-modal retrieval community have been achieved. However, existing deep cross-modal hashing methods either rely on amounts of labeled information or have no ability to learn an accuracy correlation between different modalities. In this paper, we proposed Unsupervised coupled Cycle generative adversarial Hashing networks (UCH), for cross-modal retrieval, where outer-cycle network is used to learn powerful common representation, and inner-cycle network is explained to generate reliable hash codes. Specifically, our proposed UCH seamlessly couples these two networks with generative adversarial mechanism, which can be optimized simultaneously to learn representation and hash codes. Extensive experiments on three popular benchmark datasets show that the proposed UCH outperforms the state-of-the-art unsupervised cross-modal hashing methods.
Vegetable fields in China are characterized with intensive fertilization and cultivation, and their net effect on the global warming deserves attention. Greenhouse gas fluxes were thus measured, using a static closed chamber method, over approximately 18 months in two typical subtropical vegetable fields with different soil types and contrasting soil properties. Five consecutive crops were planted in one field and four in the other. Intensive fertilization consistently stimulated soil N 2 O emission, while imposed complicated impact on soil respiration with CO 2 emission enhanced in one field and suppressed in the other field. The fertilizer-induced N 2 O emission factors (EFs) varied with individual crop phases and averaged 1.4 to 3.1% across the whole sampling period for different fields. The interaction of soil temperature and moisture could explain about 58% of the seasonal variation in the EFs. All the soils under different vegetable cropping systems were net sources of atmospheric radiative forcing and the net global warming potential over the entire study period ranged from 1,786 to 3,569 g CO 2 equivalence m -2 for fertilized soils with net CO 2 emission contributing 53 to 67% and N 2 O emission occupying the remaining 33 to 47%.
Image attribute transfer aims to change an input image to a target one with expected attributes, which has received significant attention in recent years. However, most of the existing methods lack the ability to de-correlate the target attributes and irrelevant information, i.e., the other attributes and background information, thus often suffering from blurs and artifacts. To address these issues, we propose a novel Attribute Manifold Encoding GAN (AME-GAN) for fully-featured attribute transfer, which can modify and adjust every detail in the images. Specifically, our method divides the input image into image attribute part and image background part on manifolds, which are controlled by attribute latent variables and background latent variables respectively. Through enforcing attribute latent variables to Gaussian distributions and background latent variables to uniform distributions respectively, the attribute transfer procedure becomes controllable and image generation is more photo-realistic. Furthermore, we adopt a conditional multi-scale discriminator to render accurate and high-quality target attribute images. Experimental results on three popular datasets demonstrate the superiority of our proposed method in both performances of the attribute transfer and image generation quality.
Effects of different tillage systems on organic carbon and carbon management index (CMI) in paddy soil of long-term experiment site (since 1990) were studied. The experiment included three tillage treatments: conventional tillage with rotation of rice and winter fallow (CT-r) system, no-tillage and ridge culture with rotation of rice and rape (RT-rr) system, and conventional tillage with rotation of rice and rape (CT-rr) system. Soil labile organic carbon measured by oxidation of KMnO4 respond rapidly to carbon supply changes, and it is considered as an important indicator of soil quality. Compared with CT-r system, long-term RT-rr system significantly increased total organic carbon and labile organic carbon in surface soil (0-10 cm and10-20 cm). The proportion of labile organic carbon to total organic carbon under RT-rr system was higher than other tillage systems. The carbon management index (CMI) is derived from the total soil organic carbon pool and carbon lability and is useful to evaluate the capacity of management systems to promote soil quality. The CMI increased in each layer under RT-rr system, while it decreased under CT-rr system. This indicated that conservation tillage improved the capacity of the management system into promoting soil quality in Sichuan Basin of China.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.