Zobrazeno 1 - 10
of 193
pro vyhledávání: '"Yu, Xuehui"'
Autor:
Chen, Pengfei, Xie, Lingxi, Huo, Xinyue, Yu, Xuehui, Zhang, Xiaopeng, Sun, Yingfei, Han, Zhenjun, Tian, Qi
The Segment Anything model (SAM) has shown a generalized ability to group image pixels into patches, but applying it to semantic-aware segmentation still faces major challenges. This paper presents SAM-CP, a simple approach that establishes two types
Externí odkaz:
http://arxiv.org/abs/2407.16682
Meta-Reinforcement Learning (Meta-RL) agents can struggle to operate across tasks with varying environmental features that require different optimal skills (i.e., different modes of behaviours). Using context encoders based on contrastive learning to
Externí odkaz:
http://arxiv.org/abs/2406.04815
Model-based offline Reinforcement Learning (RL) allows agents to fully utilise pre-collected datasets without requiring additional or unethical explorations. However, applying model-based offline RL to online systems presents challenges, primarily du
Externí odkaz:
http://arxiv.org/abs/2406.01065
Autor:
Yu, Xuehui, Chen, Pengfei, Wang, Kuiran, Han, Xumeng, Li, Guorong, Han, Zhenjun, Ye, Qixiang, Jiao, Jianbin
Point-based object localization (POL), which pursues high-performance object sensing under low-cost data annotation, has attracted increased attention. However, the point annotation mode inevitably introduces semantic variance due to the inconsistenc
Externí odkaz:
http://arxiv.org/abs/2401.17203
Single-point annotation in visual tasks, with the goal of minimizing labelling costs, is becoming increasingly prominent in research. Recently, visual foundation models, such as Segment Anything (SAM), have gained widespread usage due to their robust
Externí odkaz:
http://arxiv.org/abs/2312.15895
Autor:
Han, Xumeng, Wei, Longhui, Yu, Xuehui, Dou, Zhiyang, He, Xin, Wang, Kuiran, Han, Zhenjun, Tian, Qi
The recent Segment Anything Model (SAM) has emerged as a new paradigmatic vision foundation model, showcasing potent zero-shot generalization and flexible prompting. Despite SAM finding applications and adaptations in various domains, its primary lim
Externí odkaz:
http://arxiv.org/abs/2312.03628
Autor:
Cao, Guangming, Yu, Xuehui, Yu, Wenwen, Han, Xumeng, Yang, Xue, Li, Guorong, Jiao, Jianbin, Han, Zhenjun
Single-point annotation in oriented object detection of remote sensing scenarios is gaining increasing attention due to its cost-effectiveness. However, due to the granularity ambiguity of points, there is a significant performance gap between previo
Externí odkaz:
http://arxiv.org/abs/2311.13128
Object detection via inaccurate bounding boxes supervision has boosted a broad interest due to the expensive high-quality annotation data or the occasional inevitability of low annotation quality (\eg tiny objects). The previous works usually utilize
Externí odkaz:
http://arxiv.org/abs/2307.12101
Complex systems are ubiquitous in the real world and tend to have complicated and poorly understood dynamics. For their control issues, the challenge is to guarantee accuracy, robustness, and generalization in such bloated and troubled environments.
Externí odkaz:
http://arxiv.org/abs/2209.07368
Autor:
Chen, Pengfei, Yu, Xuehui, Han, Xumeng, Hassan, Najmul, Wang, Kai, Li, Jiachen, Zhao, Jian, Shi, Humphrey, Han, Zhenjun, Ye, Qixiang
Object detection using single point supervision has received increasing attention over the years. However, the performance gap between point supervised object detection (PSOD) and bounding box supervised detection remains large. In this paper, we att
Externí odkaz:
http://arxiv.org/abs/2207.06827