检索结果-内蒙古大学图书馆

arXiv 2025年

作者： Ou, Ruizhe Hu, Yuan Zhang, Fan Chen, Jiaxin Liu, Yu Pattern Recognition and Intelligent System Lab School of Artificial Intelligence Beijing University of Posts and Telecommunications Beijing100876 China Institute of Remote Sensing and Geographic Information Systems School of Earth and Space Sciences Peking University Beijing100871 China Peking University Ordos Research Institute of Energy Ordos017000 China

Multi-modal large language models (MLLMs) have achieved remarkable success in image- and region-level remote sensing (RS) image understanding tasks, such as image captioning, visual question answering, and visual grounding. However, existing RS MLLMs lack the pixel-level dialogue capability, which involves responding to user instructions with segmentation masks for specific instances. In this paper, we propose GeoPix, a RS MLLM that extends image understanding capabilities to the pixel level. This is achieved by equipping the MLLM with a mask predictor, which transforms visual features from the vision encoder into masks conditioned on the LLM’s segmentation token embeddings. To facilitate the segmentation of multi-scale objects in RS imagery, a class-wise learnable memory module is integrated into the mask predictor to capture and store class-wise geo-context at the instance level across the entire dataset. In addition, to address the absence of large-scale datasets for training pixel-level RS MLLMs, we construct the GeoPixInstruct dataset, comprising 65,463 images and 140,412 instances, with each instance annotated with text descriptions, bounding boxes, and masks. Furthermore, we develop a two-stage training strategy to balance the distinct requirements of text generation and masks prediction in multi-modal multi-task optimization. Extensive experiments verify the effectiveness and superiority of GeoPix in pixel-level segmentation tasks, while also maintaining competitive performance in image- and region-level benchmarks. The models, dataset, and code are publicly availab.e at https://***/Norman-Ou/GeoPix. © 2025, CC BY.

关键词： Image segmentation

来源：评论

学校读者我要写书评

暂无评论

ACTIVATION FORCE-BASED AIR POLLUTION TRACING 5

ACTIVATION FORCE-BASED AIR POLLUTION TRACING

引用

2016 5th IEEE International Conference on Network Infrastructure and Digital Content(IC-NIDC 2016)

作者： Huanyu Zhou Ni Zhang Di Huang Zhanyu Ma Weisong Hu Jun Guo Pattern Recognition and Intelligent System Lab. Beijing University of Posts and Telecommunications NEC Labs China

ISBN: (纸本)9781509012473

Atmospheric pollution is getting more and more attention because of the serious *** the pollutant source is one of the key steps to solve this *** paper introduces a new method to trace the pollution diffusion path according to the relevance of different areas where they were polluted(the relatively small areas regard as points or lattices in this paper).We present the single point tracing to show the specific traceability and the area tracing to show the pollution spread trend.

关键词： Air pollution Activation Force Tracing

来源：评论

学校读者我要写书评

暂无评论

An efficient algorithm of hot events detection in text streams

An efficient algorithm of hot events detection in text strea...

引用

International Conference on Cyber-Enabled Distributed Computing and Knowledge Discovery

作者： Bai, Junliang Guo, Jun Chen, Guang Xu, Weiran Du, Gang Pattern Recognition and Intelligent System Lab. Beijing University of Posts and Telecommunications China

ISBN: (纸本)9780769542355

Hot events detection in text streams has drawn increasing attention in recent sequential data mining works. Different from traditional TDT task which find all the real events' cluster, hot events detection only identify hot events concerned by public. This paper proposes a novel approach to identify those events based on burst terms, terms co-occurrence and generative probabilistic model. Experiments with huge text stream sets crawled from WWW suggest that our algorithm can work on-line and identify hot events effectively and efficiently. © 2010 IEEE.

关键词： Data mining

来源：评论

学校读者我要写书评

暂无评论

Social media communication model research bases on Sina-weibo

Social media communication model research bases on Sina-weib...

引用

作者： Wu, Ming Guo, Jun Zhang, Chuang Xie, Jianjun Pattern Recognition and Intelligent System Lab. Beijing University of Posts and Telecommunications Beijing China

ISBN: (纸本)9783642256608

The popularity of microblog brings new characters to information diffusion in social networks. Facing new challenges of understanding information propagation in microblog, the framework of information producing and receiving was proposed. A general model named competing-window is also presented based on human behavior. The detailed composition of the model and its basal mathematical description are also given. In addition, a parameter called information lost as a supplement to measure dynamics of information diffusion. Meanwhile, the further application of our model to information processing and propagating was pointed out. All those work is based on the studies on human dynamics. Finally, to verify applicability, the model was applied to empirical data crawled from Sina-weibo. The interesting patterns extracted from empirical data indicate that microblog in deed is fundamentally characterized by human dynamics. © 2011 Springer-Verlag Berlin Heidelberg.

关键词： Dynamics

来源：评论

学校读者我要写书评

暂无评论

Tweets reading probability analysis based on competing-window model

Tweets reading probability analysis based on competing-windo...

引用

International Conference on Intelligence Science and Information Engineering

作者： Xie, Jianjun Zhang, Chuang Wu, Ming Pattern Recognition and Intelligent System Lab. Beijing University of Posts and Telecommunications Beijing China

ISBN: (纸本)9780769544809

In microgblogs, a user usually follows or is followed by many other users. The content updating and reading is a complex process involving intensive interactions among publishers and readers. It also forms the basis of information diffusion in social networks. In the situation of massive followers, tweets reading would heavily depend on user behaviors and interactions. The tweets reading probability (TRP) would be a vital parameter measuring the effectiveness and influence of tweets. Our work proposed a fundamental model, namely competing-window, to simulate the process of multi-node interactions and analyzed TRP in social network. Based on Sina Microblog, we built a standard data set and run massive experiments on empirical data to extract user behavior patterns. By adopting simulating approaches, TRP in a none-preference social network was obtained. The results indicate that typical TRP is about 8% and different user behaviors affect TRP differently. © 2011 IEEE.

关键词： Probability

来源：评论

学校读者我要写书评

暂无评论

A new combination feature of S-LBP and main color descriptor for video frame retrieval

A new combination feature of S-LBP and main color descriptor...

引用

2010 2nd IEEE International Conference on Network Infrastructure and Digital Content, IC-NIDC 2010

作者： Zhang, Honggang Liu, Lixia Feng, Aiping Wan, Xinxin Guo, Jun Pattern Recognition and Intelligent System Lab. Beijing University of Posts and Telecommunications Beijing China

ISBN: (纸本)9781424468546

In this paper, we draw Local Binary pattern (LBP) and main color descriptor into content-based video indexing and retrieval. LBP is a powerful texture descriptor for its tolerance against illumination changes and its computational simplicity. Main color descriptor is used for representing main color distribution of the image. We propose a simplified LBP (S-LBP) operator as the texture descriptor. In most cases, texture distributions that ignore the actual positions or spatial arrangement in the image are not sufficient for representing an image. Therefore, we use zigzag based Run-length coding principle to produce S-LBP patterns histogram. Then we apply the combination of S-LBP feature and main color descriptor to represent an image in video retrieval. The experiment that to retrieval which video a given frame comes from demonstrates the combination method is efficient and effective. ©2010 IEEE.

关键词： Color

来源：评论

学校读者我要写书评

暂无评论

The comparison of several methods of processing no-rated items in collab.rative filtering algorithm

The comparison of several methods of processing no-rated ite...

引用

2009 2nd IEEE International Conference on Broadband Network and Multimedia Technology, IEEE IC-BNMT2009

作者： Jinbo, Zhang Zhiqing, Lin Bo, Xiao Chuang, Zhang Pattern Recognition and Intelligent System Lab. Beijing University of Posts and Telecommunications Beijing China

ISBN: (纸本)9781424450053

Collab.rative Filtering is a very important technology in E-commerce. Unfortunately, with the increase of users and commodities, the user rating data is extremely sparse, which leads to the low efficient Collab.rative Filtering recommendation system. To address these issues, many methods of processing no-rated items in Collab.rative Filtering recommendation algorithm have been proposed, including algorithm without taking the no-rated items into account, and algorithms setting the ratings of no-rated items' value as 0, half of the full score, average of the target item's rating score, or the average of the target user's rating score. This paper compares the several methods, and the experimental results show that the method of set ting no-rated items' value as 0 is the best method in these methods. ©2009 IEEE.

关键词： Collab.rative filtering

来源：评论

学校读者我要写书评

暂无评论

Simplified Local Binary pattern descriptor for character recognition of vehicle license plate

Simplified Local Binary Pattern descriptor for character rec...

引用

International Conference on Computer Graphics, Imaging and Visualization

作者： Liu, Lixia Zhang, Honggang Feng, Aiping Wan, Xinxin Guo, Jun Pattern Recognition and Intelligent System Lab. Beijing University of Posts and Telecommunications Beijing China

ISBN: (纸本)9780769541662

Local Binary pattern (LBP) is a powerful texture descriptor for its tolerance against illumination changes and its computational simplicity. The basic LBP encodes 256 feature patterns in a 3×3 neighborhood, but not all the patterns are effective for classification. In this paper, we propose a simplified LBP(S-LBP) which produces optimal patterns by using the best coding principle for classification. Meanwhile, we combine SLBP and Mahalonobis distance in solving the practical problem of character recognition in Chinese license plate. Experimental results demonstrate the effectiveness of our method for vehicle license recognition comparing with other popular methods. © 2010 IEEE.

关键词： Local binary pattern

来源：评论

学校读者我要写书评

暂无评论

BUPT at TREC 2009: Entity track

BUPT at TREC 2009: Entity track

引用

18th Text REtrieval Conference, TREC 2009

作者： Wang, Zhanyi Liu, Dongxin Xu, Weiran Chen, Guang Guo, Jun Pattern Recognition and Intelligent System Lab. Beijing University of Posts and Telecommunications Beijing 100876 China

This report introduces the work of BUPT (PRIS) in Entity Track in TREC2009. The task and data are both new this year. In our work, an improved two-stage retrieval model is proposed according to the task. The first stage is document retrieval, in order to get the similarity of the query and documents. The second stage is to find the relationship between documents and entities. We also focus on entity extraction in the second stage and the final ranking.

关键词：

来源：评论

学校读者我要写书评

暂无评论

POC-NLW template for Chinese word segmentation 5

POC-NLW template for Chinese word segmentation

引用

5th SIGHAN Workshop on Chinese Language Processing, co-located with COLING/ACL 2006

作者： Chen, Bo Xu, Weiran Peng, Tao Guo, Jun Pattern Recognition and Intelligent System Lab. Beijing University of Posts and Telecommunications Beijing100876 China

ISBN: (纸本)1932432701

In this paper, a language tagging template named POC-NLW (position of a character within an n-length word) is presented. Based on this template, a two-stage statistical model for Chinese word segmentation is constructed. In this method, the basic word segmentation is based on n-gram language model, and a Hidden Markov tagger based on the POC-NLW template is used to implement the out-of-vocabulary (OOV) word identification. The system participated in the MSRA_Close and UPUC_Close word segmentation tracks at SIGHAN Bakeoff 2006. Results returned by this bakeoff are reported here. © 2006 Association for Computational Linguistics.

关键词： Computational linguistics

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：