Zobrazeno 1 - 10
of 52 250
pro vyhledávání: '"Liu, yong"'
Boosted by Multi-modal Large Language Models (MLLMs), text-guided universal segmentation models for the image and video domains have made rapid progress recently. However, these methods are often developed separately for specific domains, overlooking
Externí odkaz:
http://arxiv.org/abs/2412.14006
Autor:
Liu, Ruyue, Yin, Rong, Bo, Xiangzhen, Hao, Xiaoshuai, Zhou, Xingrui, Liu, Yong, Ma, Can, Wang, Weiping
Federated graph learning (FGL) has gained significant attention for enabling heterogeneous clients to process their private graph data locally while interacting with a centralized server, thus maintaining privacy. However, graph data on clients are t
Externí odkaz:
http://arxiv.org/abs/2412.13442
Automatic black-and-white image sequence colorization while preserving character and object identity (ID) is a complex task with significant market demand, such as in cartoon or comic series colorization. Despite advancements in visual colorization u
Externí odkaz:
http://arxiv.org/abs/2412.11815
Autor:
Wang, Mengmeng, Ma, Teli, Xin, Shuo, Hou, Xiaojun, Xing, Jiazheng, Dai, Guang, Wang, Jingdong, Liu, Yong
Visual Object Tracking (VOT) is an attractive and significant research area in computer vision, which aims to recognize and track specific targets in video sequences where the target objects are arbitrary and class-agnostic. The VOT technology could
Externí odkaz:
http://arxiv.org/abs/2412.09991
Photon correlation is at the heart of quantum optics and has important applications in quantum technologies. Here we propose a universally applicable mechanism that can generate the superbunching light with ultrastrong second-order and higher-order c
Externí odkaz:
http://arxiv.org/abs/2412.09873
Autor:
Feng, Chun-Mei, He, Yuanyang, Zou, Jian, Khan, Salman, Xiong, Huan, Li, Zhen, Zuo, Wangmeng, Goh, Rick Siow Mong, Liu, Yong
Publikováno v:
International Journal of Computer Vision, 2025
Existing test-time prompt tuning (TPT) methods focus on single-modality data, primarily enhancing images and using confidence ratings to filter out inaccurate images. However, while image generation models can produce visually diverse images, single-
Externí odkaz:
http://arxiv.org/abs/2412.09706
Autor:
Jiao, Xianhe, Lv, Chenlei, Zhao, Junli, Yi, Ran, Wen, Yu-Hui, Pan, Zhenkuan, Wu, Zhongke, Liu, Yong-jin
For large-scale point cloud processing, resampling takes the important role of controlling the point number and density while keeping the geometric consistency. % in related tasks. However, current methods cannot balance such different requirements.
Externí odkaz:
http://arxiv.org/abs/2412.09177
Autor:
Zhang, Jiangning, Hu, Teng, He, Haoyang, Xue, Zhucun, Wang, Yabiao, Wang, Chengjie, Liu, Yong, Li, Xiangtai, Tao, Dacheng
This work focuses on developing parameter-efficient and lightweight models for dense predictions while trading off parameters, FLOPs, and performance. Our goal is to set up the new frontier of the 5M magnitude lightweight model on various downstream
Externí odkaz:
http://arxiv.org/abs/2412.06674
Autor:
Huang, Zitong, Chen, Ze, Li, Yuanze, Dong, Bowen, Zhou, Erjin, Liu, Yong, Goh, Rick Siow Mong, Feng, Chun-Mei, Zuo, Wangmeng
Few-Shot Class-Incremental Learning has shown remarkable efficacy in efficient learning new concepts with limited annotations. Nevertheless, the heuristic few-shot annotations may not always cover the most informative samples, which largely restricts
Externí odkaz:
http://arxiv.org/abs/2412.06642
Quantum criticality of open many-body systems has attracted lots of interest for emergent phenomena and universality. Here we present the exact steady state of the quantum van der Pol oscillator using the complex $P$-representation. We show the thres
Externí odkaz:
http://arxiv.org/abs/2412.03354