Zobrazeno 1 - 10
of 1 144
pro vyhledávání: '"Lai, Man On"'
Autor:
Lau, Kin Wai, Rehman, Yasar Abbas Ur, de Gusmão, Pedro Porto Buarque, Po, Lai-Man, Ma, Lan, Xie, Yuyang
Federated Learning (FL) has emerged as a privacy-preserving method for training machine learning models in a distributed manner on edge devices. However, on-device models face inherent computational power and memory limitations, potentially resulting
Externí odkaz:
http://arxiv.org/abs/2409.15898
Autor:
Po, Lai-Man, Liu, Yuyang, Wu, Haoxuan, Zhang, Tianqi, Yu, Wing-Yin, Wang, Zhuohan, Jiang, Zeyu, Li, Kun
This paper introduces Standard Basis LoRA (SBoRA), a novel parameter-efficient fine-tuning approach for Large Language Models that builds upon the pioneering works of Low-Rank Adaptation (LoRA) and Orthogonal Adaptation. SBoRA reduces the number of t
Externí odkaz:
http://arxiv.org/abs/2407.05413
Recent research has successfully adapted vision-based convolutional neural network (CNN) architectures for audio recognition tasks using Mel-Spectrograms. However, these CNNs have high computational costs and memory requirements, limiting their deplo
Externí odkaz:
http://arxiv.org/abs/2404.13551
Visual Attention Networks (VAN) with Large Kernel Attention (LKA) modules have been shown to provide remarkable performance, that surpasses Vision Transformers (ViTs), on a range of vision-based tasks. However, the depth-wise convolutional layer in t
Externí odkaz:
http://arxiv.org/abs/2309.01439
Video-based human pose transfer is a video-to-video generation task that animates a plain source human image based on a series of target human poses. Considering the difficulties in transferring highly structural patterns on the garments and disconti
Externí odkaz:
http://arxiv.org/abs/2307.07754
Autor:
Zhao, Yuzhi, Po, Lai-Man, Liu, Kangcheng, Wang, Xuehui, Yu, Wing-Yin, Xian, Pengfei, Zhang, Yujia, Liu, Mengyang
In this paper, we propose a scribble-based video colorization network with temporal aggregation called SVCNet. It can colorize monochrome videos based on different user-given color scribbles. It addresses three common issues in the scribble-based vid
Externí odkaz:
http://arxiv.org/abs/2303.11591
Night imaging with modern smartphone cameras is troublesome due to low photon count and unavoidable noise in the imaging system. Directly adjusting exposure time and ISO ratings cannot obtain sharp and noise-free images at the same time in low-light
Externí odkaz:
http://arxiv.org/abs/2207.03294
Autor:
Zhao, Yuzhi, Po, Lai-Man, Wang, Xuehui, Yan, Qiong, Shen, Wei, Zhang, Yujia, Liu, Wei, Wong, Chun-Kit, Pang, Chiu-Sing, Ou, Weifeng, Yu, Wing-Yin, Liu, Buhua
The appearances of children are inherited from their parents, which makes it feasible to predict them. Predicting realistic children's faces may help settle many social problems, such as age-invariant face recognition, kinship verification, and missi
Externí odkaz:
http://arxiv.org/abs/2204.09962
Autor:
Fung, John Tai Chun, Chan, Siu Ling, Chan, Chun Kit, Lam, Choi Fung, Chau, Yuk Shuen, Lam, Wing Hang, Cheng, Christopher Chi Wai, Lai, Man Hin
Publikováno v:
In Nurse Education Today September 2024 140
Autor:
Zhang, Yujia, Po, Lai-Man, Xu, Xuyuan, Liu, Mengyang, Wang, Yexin, Ou, Weifeng, Zhao, Yuzhi, Yu, Wing-Yin
Spatio-temporal representation learning is critical for video self-supervised representation. Recent approaches mainly use contrastive learning and pretext tasks. However, these approaches learn representation by discriminating sampled instances via
Externí odkaz:
http://arxiv.org/abs/2112.08913