Xin Xin

Xin Xin

Auto-generated profile, based on publicly available information

Xin Xin

Beijing Institute of Technology

Research Focus & Metrics

Academic Career Length

14Years (2013-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
20 November 2013 in Cognitive Computation
Verified
Journal Article
Closed access
PDF file
YYYao Yao
Yao Yao
PGPing Guo
Ping Guo
XXXin Xin
Xin Xin
ZJZiheng Jiang
Ziheng Jiang

8 July 2015 in Journal of Computer Science and Technology
Verified
Journal Article
Closed access
PDF file
XXXin Xin
Xin Xin
CLChin-Yew Lin
Chin-Yew Lin
XWXiao-Chi Wei
Xiao-Chi Wei
HHHe-Yan Huang
He-Yan Huang

20 May 2015 in Cognitive Computation
Verified
Journal Article
Open access
PDF file
YYYao Yao
Yao Yao
PGPing Guo
Ping Guo
XXXin Xin
Xin Xin
ZJZiheng Jiang
Ziheng Jiang

9 December 2015
Verified
Book Chapter
Closed access
PDF file
YLYanjun Li
Yanjun Li
XXXin Xin
Xin Xin
PGPing Guo
Ping Guo

18 December 2024 in arXiv
Verified
Preprint
Open access
PDF file
ZXZhengfei Xu
Zhengfei Xu
SZSijia Zhao
Sijia Zhao
YHYanchao Hao
Yanchao Hao
XLXiaolong Liu
Xiaolong Liu
LLLili Li
Lili Li
XXXin Xin
Xin Xin
Visual Entity Linking (VEL) is a crucial task for achieving fine-grained visual understanding, matching objects within images (visual mentions) to entities in a knowledge base. Previous VEL tasks rely on textual inputs, but writing queries for complex scenes can be challenging. Visual inputs like clicks or bounding boxes offer a more convenient alternative. Therefore, we propose a new task, Pixel-Level Visual Entity Linking (PL-VEL), which uses pixel masks from visual inputs to refer to objects, supplementing reference methods for VEL. To facilitate research on this task, we have constructed the MaskOVEN-Wiki dataset through an entirely automatic reverse region-entity annotation framework. This dataset contains over 5 million annotations aligning pixel-level regions with entity-level labels, which will advance visual understanding towards fine-grained. Moreover, as pixel masks correspond to semantic regions in an image, we enhance previous patch-interacted attention with region-interacted attention by a visual semantic tokenization approach. Manual evaluation results indicate that the reverse annotation framework achieved a 94.8% annotation success rate. Experimental results show that models trained on this dataset improved accuracy by 18 points compared to zero-shot models. Additionally, the semantic tokenization method achieved a 5-point accuracy improvement over the trained baseline.

11 April 2025 in Proceedings of the AAAI Conference on Artificial Intelligence
Verified
Journal Article
Open access
PDF file
ZXZhengfei Xu
Zhengfei Xu
SZSijia Zhao
Sijia Zhao
YHYanchao Hao
Yanchao Hao
XLXiaolong Liu
Xiaolong Liu
LLLili Li
Lili Li
XXXin Xin
Xin Xin
Visual Entity Linking (VEL) is a crucial task for achieving fine-grained visual understanding, matching objects within images (visual mentions) to entities in a knowledge base. Previous VEL tasks rely on textual inputs, but writing queries for complex scenes can be challenging. Visual inputs like clicks or bounding boxes offer a more convenient alternative. Therefore, we propose a new task, Pixel-Level Visual Entity Linking (PL-VEL), which uses pixel masks from visual inputs to refer to objects, supplementing reference methods for VEL. To facilitate research on this task, we have constructed the MaskOVEN-Wiki dataset through an entirely automatic reverse region-entity annotation framework. This dataset contains over 5 million annotations aligning pixel-level regions with entity-level labels, which will advance visual understanding towards fine-grained. Moreover, as pixel masks correspond to semantic regions in an image, we enhance previous patch-interacted attention with region-interacted attention by a visual semantic tokenization approach. Manual evaluation results indicate that the reverse annotation framework achieved a 94.8% annotation success rate. Experimental results show that models trained on this dataset improved accuracy by 18 points compared to zero-shot models. Additionally, the semantic tokenization method achieved a 5-point accuracy improvement over the trained baseline.