Zobrazeno 1 - 10
of 2 294
pro vyhledávání: '"DENG Rui"'
The success of speech-image retrieval relies on establishing an effective alignment between speech and image. Existing methods often model cross-modal interaction through simple cosine similarity of the global feature of each modality, which fall sho
Externí odkaz:
http://arxiv.org/abs/2408.13705
Autor:
Deng, Rui, Gu, Tianpei
Reconstructing degraded images is a critical task in image processing. Although CNN and Transformer-based models are prevalent in this field, they exhibit inherent limitations, such as inadequate long-range dependency modeling and high computational
Externí odkaz:
http://arxiv.org/abs/2404.11778
Differentiable Resolution Compression and Alignment for Efficient Video Classification and Retrieval
Optimizing video inference efficiency has become increasingly important with the growing demand for video analysis in various fields. Some existing methods achieve high efficiency by explicit discard of spatial or temporal information, which poses ch
Externí odkaz:
http://arxiv.org/abs/2309.08167
Training Graph Neural Networks (GNNs) on real-world graphs consisting of billions of nodes and edges is quite challenging, primarily due to the substantial memory needed to store the graph and its intermediate node and edge features, and there is a p
Externí odkaz:
http://arxiv.org/abs/2308.03209
Publikováno v:
Feed Research. 2024, Vol. 47 Issue 18, p26-29. 4p.
This paper introduces the system submitted by dun_oscar team for the ICPR MSR Challenge. Three subsystems for task1-task3 are descripted respectively. In task1, we develop a visual system which includes a OCR model, a text tracker, and a NLP classifi
Externí odkaz:
http://arxiv.org/abs/2303.06878
Autor:
Deng, Rui1,2 (AUTHOR) d434314909@gmail.com, Wang, Mingshuai3 (AUTHOR) 20720191150071@stu.xmu.edu.cn, Zhang, Hao1,2 (AUTHOR) 23s136528@stu.hit.edu.cn, Yao, Ruijun1 (AUTHOR) cumtyrj1996@126.com, Zhen, Kai1,2 (AUTHOR) zhkaihit@163.com, Liu, Yifei1,2 (AUTHOR) l1242070761@163.com, Liu, Xingjun1,2,3 (AUTHOR) xjliu@hit.edu.cn, Wang, Cuiping3 (AUTHOR) xjliu@hit.edu.cn
Publikováno v:
Materials (1996-1944). Oct2024, Vol. 17 Issue 20, p4973. 12p.
In this paper, we introduce 3D-CSL, a compact pipeline for Near-Duplicate Video Retrieval (NDVR), and explore a novel self-supervised learning strategy for video similarity learning. Most previous methods only extract video spatial features from fram
Externí odkaz:
http://arxiv.org/abs/2211.05352
Autor:
DENG Rui-yun, WANG Kun-yao, REN Xin-ru LIU Rong-jun, CUI Zhi-jing, ZENG Fan-li, LI Li, WANG Zhen
Publikováno v:
Feed Research. 2024, Vol. 47 Issue 9, p113-117. 5p.
Publikováno v:
In Industrial Crops & Products 15 December 2024 222 Part 3