Zobrazeno 1 - 10
of 102
pro vyhledávání: '"Lv, Fengmao"'
Self-supervised learning (SSL) has shown remarkable data representation capabilities across a wide range of datasets. However, when applied to real-world datasets with long-tailed distributions, performance on multiple downstream tasks degrades signi
Externí odkaz:
http://arxiv.org/abs/2410.22883
Autor:
Zou, Heqing, Luo, Tianze, Xie, Guiyang, Victor, Zhang, Lv, Fengmao, Wang, Guangcong, Chen, Juanyang, Wang, Zhuochen, Zhang, Hansheng, Zhang, Huaijian
The integration of Large Language Models (LLMs) with visual encoders has recently shown promising performance in visual understanding tasks, leveraging their inherent capability to comprehend and generate human-like text for visual reasoning. Given t
Externí odkaz:
http://arxiv.org/abs/2409.18938
Semi-supervised multi-label feature selection has recently been developed to solve the curse of dimensionality problem in high-dimensional multi-label data with certain samples missing labels. Although many efforts have been made, most existing metho
Externí odkaz:
http://arxiv.org/abs/2406.12193
Facial action unit (AU) detection has long encountered the challenge of detecting subtle feature differences when AUs activate. Existing methods often rely on encoding pixel-level information of AUs, which not only encodes additional redundant inform
Externí odkaz:
http://arxiv.org/abs/2402.06165
Colorectal polyp segmentation (CPS), an essential problem in medical image analysis, has garnered growing research attention. Recently, the deep learning-based model completely overwhelmed traditional methods in the field of CPS, and more and more de
Externí odkaz:
http://arxiv.org/abs/2401.11734
Although multi-view unsupervised feature selection (MUFS) is an effective technology for reducing dimensionality in machine learning, existing methods cannot directly deal with incomplete multi-view data where some samples are missing in certain view
Externí odkaz:
http://arxiv.org/abs/2401.10549
Multimodal sentiment analysis aims to identify the emotions expressed by individuals through visual, language, and acoustic cues. However, most of the existing research efforts assume that all modalities are available during both training and testing
Externí odkaz:
http://arxiv.org/abs/2401.10747
In the era of big data, the issue of data quality has become increasingly prominent. One of the main challenges is the problem of duplicate data, which can arise from repeated entry or the merging of multiple data sources. These "dirty data" problems
Externí odkaz:
http://arxiv.org/abs/2308.00721
Autor:
Huang, Yanyong, Shen, Zongxin, Cai, Yuxin, Yi, Xiuwen, Wang, Dongjie, Lv, Fengmao, Li, Tianrui
Multi-view unsupervised feature selection (MUFS) has been demonstrated as an effective technique to reduce the dimensionality of multi-view unlabeled data. The existing methods assume that all of views are complete. However, multi-view data are usual
Externí odkaz:
http://arxiv.org/abs/2208.09736
Publikováno v:
In Image and Vision Computing November 2024 151
