Search

Your search keyword '"Zhang, Xudong"' showing total 2,113 results
Start Over Author "Zhang, Xudong" Search Limiters Peer Reviewed
2,113 results on '"Zhang, Xudong"'

22. MMFL-Net: Multi-scale and Multi-granularity Feature Learning for Cross-domain Fashion Retrieval

Author

Bao, Chen, Zhang, Xudong, Chen, Jiazhou, and Miao, Yongwei

Subjects
Computer Science - Computer Vision and Pattern Recognition, Computer Science - Artificial Intelligence
Abstract

Instance-level image retrieval in fashion is a challenging issue owing to its increasing importance in real-scenario visual fashion search. Cross-domain fashion retrieval aims to match the unconstrained customer images as queries for photographs provided by retailers; however, it is a difficult task due to a wide range of consumer-to-shop (C2S) domain discrepancies and also considering that clothing image is vulnerable to various non-rigid deformations. To this end, we propose a novel multi-scale and multi-granularity feature learning network (MMFL-Net), which can jointly learn global-local aggregation feature representations of clothing images in a unified framework, aiming to train a cross-domain model for C2S fashion visual similarity. First, a new semantic-spatial feature fusion part is designed to bridge the semantic-spatial gap by applying top-down and bottom-up bidirectional multi-scale feature fusion. Next, a multi-branch deep network architecture is introduced to capture global salient, part-informed, and local detailed information, and extracting robust and discrimination feature embedding by integrating the similarity learning of coarse-to-fine embedding with the multiple granularities. Finally, the improved trihard loss, center loss, and multi-task classification loss are adopted for our MMFL-Net, which can jointly optimize intra-class and inter-class distance and thus explicitly improve intra-class compactness and inter-class discriminability between its visual representations for feature learning. Furthermore, our proposed model also combines the multi-task attribute recognition and classification module with multi-label semantic attributes and product ID labels. Experimental results demonstrate that our proposed MMFL-Net achieves significant improvement over the state-of-the-art methods on the two datasets, DeepFashion-C2S and Street2Shop., Comment: 27 pages, 12 figures, Published by

Published
2022
Full Text
View/download PDF

32. Water, nitrogen, and phosphorus coupling improves gray jujube fruit quality and yield

Author

Zhang Xudong, Wang Jingjing, Bai Xinlu, An Shijie, Zheng Qiangqing, Tang Zhihui, and Zhi Jinhu

Subjects
gray jujube, water and fertilizer integration, fruit quality, yield, comprehensive evaluation, yield model, economic model, Biology (General), QH301-705.5
Abstract

Irrigation and fertilization are indispensable links in the jujube planting industry in southern Xinjiang, China. Regulating the relationship between fertilization and irrigation can effectively reduce costs and improve economic efficiency. A 2-year water and fertilizer optimization coupling test was conducted to determine the optimal water and nutrient supply scheme. The three-factor randomized block experiment included water (W), nitrogen (N), and phosphorus (P). According to the principal component analysis of each index, each treatment’s comprehensive score was obtained. Using yield and economic regression models, the theoretical value and yield value of the optimal economic benefit are inferred. When W, N, and P were applied together, the fruit quality and yield of each treatment significantly differed, and the vitamin C, soluble sugar, and sugar-acid ratio increased significantly with an increase in N fertilizer. However, the titratable acid decreased. An increase in irrigation and nitrogen application significantly increased fruit yield. The comprehensive score was the highest in the N4P3W2 treatment, which improved fruit quality, and the lowest in the N3P3W2 treatment. When the amounts of N, P, and W were 275.56 kg hm−2, 413. 66 kg hm−2, and 7278.19 m3 hm−2, respectively, the theoretical economic benefit was the best. The N4P3W2 treatment is the optimal treatment.

Published
2024
Full Text
View/download PDF

Catalog

Books, media, physical & digital resources
See catalog results