检索结果-内蒙古大学图书馆

10th International Conference on Virtual Reality, ICVR 2024

作者： Hu, Tianrui Hou, Wenjun School of Digital Media & Design Arts Beijing China Beijing Key Laboratory of Network System and Network Culture China Ministry of Culture and Tourism Key Laboratory of Interactive Technology and Experience System China

ISBN: (纸本)9798350364231

Subtitles (captions displayed on the screen) play a crucial role in cross-lingual multimedia content, serving as valuable aids for individuals who are deaf or hearing-impaired, as well as in situations involving unknown languages or loud environments. Subtitles are equally indispensable when viewing 360° immersive videos in virtual reality (VR). However, the experience of watching 360° videos in VR differs significantly from that on desktop screens. Hence, we propose a method for presenting VR 360° video subtitles that leverages users' gaze data as indirect input to dynamically adjust subtitle positions in real-time. Our research investigates the impact of adaptive subtitles compared to traditional subtitles on users' experiences when viewing 360° videos in VR *** findings suggest that the proposed adaptive subtitle placement method is more effective than traditional subtitles located below the field of view in reducing visual fatigue and enhancing the watching experience for users. Furthermore, in comparison to conversational 360° videos featuring speaking subjects, adaptive subtitles offer greater optimization and improvement in scene-oriented videos with voice-overs. © 2024 IEEE.

关键词： Virtual environments

来源：评论

学校读者我要写书评

暂无评论

Lulu Timeline: VR Video Timeline with Expanded interactive Scope and Fine-tuning Function 10

Lulu Timeline: VR Video Timeline with Expanded Interactive S...

引用

10th International Conference on Virtual Reality, ICVR 2024

作者： Yu, Jiaao Hou, Wenjun School of Digital Media & Design Arts Beijing China Beijing Key Laboratory of Network System and Network Culture China Ministry of Culture and Tourism Key Laboratory of Interactive Technology and Experience System China

ISBN: (纸本)9798350364231

With the growing market of virtual reality(VR) during these years, watching immersive videos in VR headset has become more and more popular, which demands more convenient video player design in immersive environment. Timeline control, as an important part of video player, enabling users to seek to particular time of video, however has no mature design established in VR yet. We propose Lulu Timeline, a VR video timeline design to enhance users' ability of controlling video play time and reduce work load, which utilizes space that is often overlooked in general design, and would adjust the scaling of timeline playhead movement based on video duration. To determine the actual effect of this new timeline, we conducted a user study with 19 participants, in which we compared our design with the most popular design currently within two tasks. Based on the analysis of subjective and objective data, our design performs higher accuracy, speed, and can significantly reduce users' workload, helping them focus their attention on areas-of-interest. © 2024 IEEE.

关键词： Virtual environments

来源：评论

学校读者我要写书评

暂无评论

MotionMLP: End-To-End Action Recognition with Motion Aware Vision MLP

MotionMLP: End-To-End Action Recognition with Motion Aware V...

引用

2023 IEEE International Conference on Visual Communications and Image Processing, VCIP 2023

作者： Ruan, Xiangning Zhao, Zhicheng Su, Fei Beijing University of Posts and Telecommunications China Beijing Key Laboratory of Network System and Network Culture China Ministry of Culture and Tourism Key Laboratory of Interactive Technology and Experience System China

ISBN: (纸本)9798350359855

Action recognition aims to interpret complex spatiotemporal patterns in the video. Current methods utilize CNN or Transformer structures, requiring extensive pre-Training methods and optical flow to capture motion information. Such approaches are computationally expensive, necessitate significant storage, cannot be trained end-To-end, and typically neglect joint learning of temporal and spatial streams. In this paper, we propose MotionMLP, a novel MLP architecture that extracts motion information from videos, then dynamically adjusts the connection between tokens and static weights within the MLP structure. The MotionMLP solely relies on video frames as input and is independent of any pre-Training method or optical flow computation. The experimental results indicate that MotionMLP outperforms the previous SOTA real-Time end-To-end methods on UCF101 and HMDB51, and relies on one-Tenth of the parameters compared with typical two-stream CNN approaches, while operating ten times faster. © 2023 IEEE.

关键词： Optical flows

来源：评论

学校读者我要写书评

暂无评论

Filter or Compensate: Towards Invariant Representation from Distribution Shift for Anomaly Detection 39

Filter or Compensate: Towards Invariant Representation from ...

引用

39th Annual AAAI Conference on Artificial Intelligence, AAAI 2025

作者： Chen, Zining Luo, Xingshuang Wang, Weiqiu Zhao, Zhicheng Su, Fei Men, Aidong School of Artificial Intelligence Beijing University of Posts and Telecommunications China Beijing Key Laboratory of Network System and Network Culture China Key Laboratory of Intereactive Technology and Experience System Ministry of Culture and Tourism Beijing China

ISBN: (纸本)157735897X

Recent Anomaly Detection (AD) methods have achieved great success with In-Distribution (ID) data. However, real-world data often exhibits distribution shift, causing huge performance decay on traditional AD methods. From this perspective, few previous work has explored AD with distribution shift, and the distribution-invariant normality learning has been proposed based on the Reverse Distillation (RD) framework. However, we observe the misalignment issue between the teacher and the student network that causes detection failure, thereby propose FiCo, Filter or Compensate, to address the distribution shift issue in AD. FiCo firstly compensates the distribution-specific information to reduce the misalignment between the teacher and student network via the Distribution-Specific Compensation (DiSCo) module, and secondly filters all abnormal information to capture distribution-invariant normality with the Distribution-Invariant Filter (DiIFi) module. Extensive experiments on three different AD benchmarks demonstrate the effectiveness of FiCo, which outperforms all existing state-of-the-art (SOTA) methods, and even achieves better results on the ID scenario compared with RD-based methods. © 2025, Association for the Advancement of Artificial Intelligence (***). All rights reserved.

关键词： Normal distribution

来源：评论

学校读者我要写书评

暂无评论

Communicative MARL-based Relevance Discerning Network for Repetition-Aware Recommendation 23

Communicative MARL-based Relevance Discerning Network for Re...

引用

32nd ACM World Wide Web Conference, WWW 2023

作者： Li, Kaiyuan Wang, Pengfei Wang, Haitao Liu, Qiang Wang, Xingxing Wang, Dong Wang, Shangguang Beijing University of Posts and Telecommunications Beijing China Meituan Beijing China Key Laboratory of Interactive Technology and Experience System Ministry of Culture and Tourism Beijing China

ISBN: (纸本)9781450394161

The repeated user-item interaction now is becoming a common phenomenon in the e-commerce scenario. Due to its potential economic profit, various models are emerging to predict which item will be re-interacted based on the user-item interactions. In this specific scenario, item relevance is a critical factor that needs to be concerned, which tends to have different effects on the succeeding re-interacted one (i.e., stimulating or delaying its emergence). It is necessary to make a detailed discernment of item relevance for a better repetition-aware recommendation. Unfortunately, existing works usually mixed all these types, which may disturb the learning process and result in poor performance. In this paper, we introduce a novel Communicative MARL-based Relevance Discerning Network (CARDfor short) to automatically discern the item relevance for a better repetition-aware recommendation. Specifically, CARDformalizes the item relevance discerning problem into a communication selection process in MARL. CARDtreats each unique interacted item as an agent and defines three different communication types over agents, which are stimulative, inhibitive, and noisy respectively. After this, CARDutilizes a Gumbel-enhanced classifier to distinguish the communication types among agents, and an attention-based Reactive Point Process is further designed to transmit the well-discerned stimulative and inhibitive incentives separately among all agents to make an effective collaboration for repetition decisions. Experimental results on two real-world e-commerce datasets show that our proposed method outperforms the state-of-the-art recommendation methods in terms of both sequential and repetition-aware recommenders. Furthermore, CARDis also deployed in the online sponsored search advertising system in Meituan, obtaining a performance improvement of over 1.5% and 1.2% in CTR and effective Cost Per Mille (eCPM) respectively, which is significant to the business. © 2023 ACM.

关键词： Multi agent systems

来源：评论

学校读者我要写书评

暂无评论

UFO: Enhancing Diffusion-Based Video Generation with a Uniform Frame Organizer 39

UFO: Enhancing Diffusion-Based Video Generation with a Unifo...

引用

39th Annual AAAI Conference on Artificial Intelligence, AAAI 2025

作者： Liu, Delong Hou, Zhaohui Zhan, Mingjie Han, Shihao Zhao, Zhicheng Su, Fei School of Artificial Intelligence Beijing University of Posts and Telecommunications China SenseTime China Beijing Key Laboratory of Network System and Network Culture China Key Laboratory of Intereactive Technology and Experience System Ministry of Culture and Tourism Beijing China

ISBN: (纸本)157735897X

Recently, diffusion-based video generation models have achieved significant success. However, existing models often suffer from issues like weak consistency and declining image quality over time. To overcome these challenges, inspired by aesthetic principles, we propose a non-invasive plug-in called Uniform Frame Organizer (UFO), which is compatible with any diffusion-based video generation model. The UFO comprises a series of adaptive adapters with adjustable intensities, which can significantly enhance the consistency between the foreground and background of videos and improve image quality without altering the original model parameters when integrated. The training for UFO is simple, efficient, requires minimal resources, and supports stylized training. Its modular design allows for the combination of multiple UFOs, enabling the customization of personalized video generation models. Furthermore, the UFO also supports direct transferability across different models of the same specification without the need for specific retraining. The experimental results indicate that UFO effectively enhances video generation quality and demonstrates its superiority in public video generation benchmarks. © 2025, Association for the Advancement of Artificial Intelligence (***). All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Research on design thinking workshop based on protocol analysis: A case study of library usage scenarios

Advanced Design Research

引用

Advanced Design Research 2024年第1期2卷 63-75页

作者： Qing Sheng Nan Liu Chenlu Zhou Suhan Li Zihao Wang Beijing Key Laboratory of Network System and Network Culture Beijing University of Posts and Telecommunications China Key Laboratory of Interactive Technology and Experience System Ministry of Culture and Tourism China

Design thinking has a broad range of applications beyond the design field, such as in libraries. However, significant differences remain between designers and non-designers in using design thinking tools for innovation. This study investigated these distinctions by examining the design thinking process. In a design thinking workshop focused on library usage, we recruited three groups of participants consisting of both designers and non-designers, who were assigned the same tasks and conducted similar activities. Through protocol analysis, significant cognitive differences between designers and non-designers in the design thinking process were found to involve information intention, long-term memory, working memory, creative thinking, and externalization. We concluded that the involvement of a designer is optimal for non-designers when using design thinking for innovation, particularly during the problem-framing stage. Additionally, we summarized the challenges faced by non-designers in utilizing design thinking for innovation and provided recommendations for improving design thinking workshops, including considerations for time management and staff allocation.

关键词： Protocol analysis Design thinking Workshop Non-designer Designer

来源：评论

学校读者我要写书评

暂无评论

Multi-Scale Context Aggregation Network for Inharmonious Region Localization 15

Multi-Scale Context Aggregation Network for Inharmonious Reg...

引用

15th IEEE International Conference on Advanced Infocomm technology, ICAIT 2023

作者： Zhang, Shu Huang, Hai Zhu, Yueyan Beijing University of Posts and Telecommunications School of Information and Communication Engineering Beijing100876 China Ministry of Culture and Tourism Key Laboratory of Interactive Technology and Experience System Beijing100876 China

ISBN: (纸本)9798350314120

With the ubiquity of image editing techniques, synthetic images produced by users often suffer from a lack of visual coherence due to color and illumination discrepancies between foreground objects and their surrounding backgrounds. Inharmonious region localization (IRL) naturally arises as a direction for exploration following image harmonization. As a fundamental task, inharmonious region localization aims at identifying the inharmonious regions in an image. In this paper, we propose a multi-scale context aggregation network for inharmonious region localization. Specifically, we employ a U-like structure to extract multi-scale features and adopt polarized self-attention mechanism for cross-scale context aggregation. In addition, we design a focal loss to balance the positive and negative samples. We compare our proposed method with existing IRL methods on benchmark datasets. Experimental results show that our method achieves significantly better performance, which demonstrates the effectiveness of our method. © 2023 IEEE.

关键词： Computer vision

来源：评论

学校读者我要写书评

暂无评论

Enhanced Machine Reading Comprehension Method for Aspect Sentiment Quadruplet Extraction 26

Enhanced Machine Reading Comprehension Method for Aspect Sen...

引用

26th European Conference on Artificial Intelligence, ECAI 2023

作者： Ye, Shuqin Zhai, Zepeng Li, Ruifan School of Artificial Intelligence Beijing University of Posts and Telecommunications China Engineering Research Center of Information Networks Ministry of Education China Key Laboratory of Interactive Technology and Experience System Ministry of Culture and Tourism China

ISBN: (纸本)9781643684369

In the NLP domain, Aspect-Based Sentiment Analysis (ABSA) has gained significant attention in recent years due to its ability to perform fine-grained sentiment analysis. A challenging task in ABSA is Aspect Sentiment Quadruplet Extraction (ASQE), which involves the extraction of aspect terms and their associated opinion terms, sentiment polarities, and categories in the form of quadruplets. However, existing studies have ignored the strong dependence among the multiple subtasks involved in ASQE. In this paper, we propose a novel Enhanced Machine Reading Comprehension (EMRC) method and formalize ASQE task as a multi-turn MRC task. Our EMRC effectively learns and utilizes the relationships among different subtasks by incorporating previously generated query answers into the current queries. We design a hierarchical category classification strategy to perform the category prediction in a structured manner, enabling the model to tackle intricate categories with ease. Furthermore, we employ the bi-directional attention mechanism, i.e., context-to-query and query-to-context attentions, to map the context into a task-aware representation. We conduct extensive experiments on two benchmark datasets. The results demonstrate that EMRC outperforms the state-of-art baselines. The source code is publicly available at https://***/Little-Yeah/EMCR. © 2023 The Authors.

关键词： Extraction

来源：评论

学校读者我要写书评

暂无评论

Multimodal Aspect-Based Sentiment Analysis under Conditional Relation 31

Multimodal Aspect-Based Sentiment Analysis under Conditional...

引用

31st International Conference on Computational Linguistics, COLING 2025

作者： Liu, Xinjing Li, Ruifan Ye, Shuqin Zhang, Guangwei Wang, Xiaojie School of Artificial Intelligence Beijing University of Posts and Telecommunications China School of Computer Science Beijing University of Posts and Telecommunications China Engineering Research Center of Information Networks Ministry of Education China Key Laboratory of Interactive Technology and Experience System Ministry of Culture and Tourism China

ISBN: (纸本)9798891761964

Multimodal Aspect-Based Sentiment Analysis (MABSA) aims to extract aspect terms from text-image pairs and identify their sentiments. Previous methods are based on the premise that the image contains the objects referred by the aspects within the text. However, this condition cannot always be met, resulting in a suboptimal performance. In this paper, we propose COnditional Relation based Sentiment Analysis framework (CORSA). Specifically, we design a conditional relation detector (CRD) to mitigate the impact of the unmet conditional image. Moreover, we design a visual object localizer (VOL) to locate the exact condition-related visual regions associated with the aspects. With CRD and VOL, our CORSA framework takes a multi-task form. In addition, to effectively learn CORSA we conduct two types of annotations. One is the conditional relation using a pretrained referring expression comprehension model;the other is the bounding boxes of visual objects by a pretrained object detection model. Experiments on our built C-MABSA dataset show that CORSA consistently outperforms existing methods. The code and data are available at https://***/Liuxj-Anya/CORSA. © 2025 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：