检索结果-内蒙古大学图书馆

TC-GS: Tri-plane based Compression for 3D Gaussian Splatting

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Wang, Taorui Yu, Zitong Xu, Yong Shenzhen Key Laboratory of Visual Object Detection and Recognition Harbin Institute of Technology Shenzhen China School of Computing and Information Technology Great Bay University China Dongguan Key Laboratory for Intelligence and Information Technology China

Recently, 3D Gaussian Splatting (3DGS) has emerged as a prominent framework for novel view synthesis, providing high fidelity and rapid rendering speed. However, the substantial data volume of 3DGS and its attributes impede its practical utility, requiring compression techniques for reducing memory cost. Nevertheless, the unorganized shape of 3DGS leads to difficulties in compression. To formulate unstructured attributes into normative distribution, we propose a well-structured tri-plane to encode Gaussian attributes, leveraging the distribution of attributes for compression. To exploit the correlations among adjacent Gaussians, K-Nearest Neighbors (KNN) is used when decoding Gaussian distribution from the Triplane. We also introduce Gaussian position information as a prior of the position-sensitive decoder. Additionally, we incorporate an adaptive wavelet loss, aiming to focus on the high-frequency details as iterations increase. Our approach has achieved results that are comparable to or surpass that of SOTA 3D Gaussians Splatting compression work in extensive experiments across multiple datasets. The codes is available at TC-GS. Copyright © 2025, The Authors. All rights reserved.

关键词： Nearest neighbor search

Pix2Cap-COCO: Advancing visual Comprehension via Pixel-Level Captioning

学校读者我要写书评

暂无评论

arXiv 2025年

作者： You, Zuyao Wang, Junke Kong, Lingyu He, Bo Wu, Zuxuan Shanghai Key Lab of Intell. Info. Processing School of CS Fudan University China Shanghai Collaborative Innovation Center of Intelligent Visual Computing China University of Maryland College Park United States

(Image Present) We present Pix2Cap-COCO, the first panoptic pixel-level caption dataset designed to advance fine-grained visual understanding. To achieve this, we carefully design an automated annotation pipeline that prompts GPT-4V to generate pixel-aligned, instance-specific captions for individual objects within images, enabling models to learn more granular relationships between objects and their contexts. This approach results in 167,254 detailed captions, with an average of 22.94 words per caption. Building on Pix2Cap-COCO, we introduce a novel task, panoptic segmentation-captioning, which challenges models to recognize instances in an image and provide detailed descriptions for each simultaneously. To benchmark this task, we design a robust baseline based on X-Decoder. The experimental results demonstrate that Pix2Cap-COCO is a particularly challenging dataset, as it requires models to excel in both fine-grained visual understanding and detailed language generation. Furthermore, we leverage Pix2Cap-COCO for Supervised Fine-Tuning (SFT) on large multimodal models (LMMs) to enhance their performance. For example, training with Pix2Cap-COCO significantly improves the performance of GPT4RoI, yielding gains in CIDEr (+1.4%), ROUGE (+0.4%), and SPICE (+0.5%) on visual Genome dataset, and strengthens its region understanding ability on the ViP-Bench, with an overall improvement of +5.1%, including notable increases in recognition accuracy (+11.2%) and language generation quality (+22.2%). Code is available at https://***/geshang777/pix2cap. © 2025, CC BY-NC-SA.

关键词： Pixels

EDEN: Enhanced Diffusion for High-quality Large-motion Video Frame Interpolation

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Zhang, Zihao Chen, Haoran Zhao, Haoyu Lu, Guansong Fu, Yanwei Xu, Hang Wu, Zuxuan Shanghai Key Lab of Intell. Info. Processing School of CS Fudan University China Shanghai Collaborative Innovation Center of Intelligent Visual Computing China Noah’s Ark Lab Huawei Canada

Handling complex or nonlinear motion patterns has long posed challenges for video frame interpolation. Although recent advances in diffusion-based methods offer improvements over traditional optical flow-based approaches, they still struggle to generate sharp, temporally consistent frames in scenarios with large motion. To address this limitation, we introduce EDEN, an Enhanced Diffusion for high-quality large-motion vidEo frame iNterpolation. Our approach first utilizes a transformer-based tokenizer to produce refined latent representations of the intermediate frames for diffusion models. We then enhance the diffusion transformer with temporal attention across the process and incorporate a start-end frame difference embedding to guide the generation of dynamic motion. Extensive experiments demonstrate that EDEN achieves state-of-the-art results across popular benchmarks, including nearly a 10% LPIPS reduction on DAVIS and SNU-FILM, and an 8% improvement on DAIN-HD. Copyright © 2025, The Authors. All rights reserved.

关键词： Optical flows

Beyond Time and Accuracy: Strategies in visual Problem-Solving 25

学校读者我要写书评

暂无评论

Beyond Time and Accuracy: Strategies in Visual Problem-Solvi...

Proceedings of the 2025 CHI Conference on Human Factors in computing Systems

作者： Eric Mörth Zona Kostic Nils Gehlenborg Hanspeter Pfister Johanna Beyer Carolina Nobre Department of Biomedical Informatics Harvard Medical School Boston Massachusetts USA SEAS Harvard University Boston Massachusetts USA Visual Computing Group Harvard University Cambridge Massachusetts USA University of Toronto Toronto Ontario Canada

Named Entity Recognition for Serbian Legal Documents: Design, Methodology and Dataset Development

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Kalušev, Vladimir Brkljač, Branko Visual Computing & Perception Group The Institute for Artificial Intelligence Research and Development of Serbia Novi Sad Serbia Dept. of Power Electronic and Telecommunication Engineering Faculty of Technical Sciences University of Novi Sad Novi Sad Serbia

Recent advancements in the field of natural language processing (NLP) and especially large language models (LLMs) and their numerous applications have brought research attention to design of different document processing tools and enhancements in the process of document archiving, search and retrieval. Domain of official, legal documents is especially interesting due to vast amount of data generated on the daily basis, as well as the significant community of interested practitioners (lawyers, law offices, administrative workers, state institutions and citizens). Providing efficient ways for automation of everyday work involving legal documents is therefore expected to have significant impact in different fields. In this work we present one LLM based solution for Named Entity Recognition (NER) in the case of legal documents written in Serbian language. It leverages on the pre-trained bidirectional encoder representations from transformers (BERT), which had been carefully adapted to the specific task of identifying and classifying specific data points from textual content. Besides novel dataset development for Serbian language (involving public court rulings), presented system design and applied methodology, the paper also discusses achieved performance metrics and their implications for objective assessment of the proposed solution. Performed cross-validation tests on the created manually labeled dataset with mean F1 score of 0.96 and additional results on the examples of intentionally modified text inputs confirm applicability of the proposed system design and robustness of the developed NER *** Codes 68T10, 68T30, 68T35, 68T50, 91F20 © 2025, CC BY.

关键词： Natural language processing systems

XML3D and Xflow Combining Declarative 3D for the Web with Generic Data Flows

学校读者我要写书评

暂无评论

IEEE COMPUTER GRAPHICS AND APPLICATIONS 2013年第5期33卷 38-47页

作者： Klein, Felix Sons, Kristian Rubinstein, Dmitri Slusallek, Philipp Intel Visual Computing Institute|c|

Researchers have combined XML3D, which provides declarative, interactive 3D scene descriptions based on HTML5, with Xflow, a language for declarative, high-performance data processing. The result lets Web developers c... 详细信息

关键词： River Trail Web3D WebGL XML3D Xflow computer graphics data processing parallel JavaScript

EventEgo3D++: 3D Human Motion Capture from a Head-Mounted Event Camera

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Millerdurai, Christen Akada, Hiroyasu Wang, Jian Luvizon, Diogo Pagani, Alain Stricker, Didier Theobalt, Christian Golyanik, Vladislav Visual Computing and Artificial Intelligence Max Planck Institute for Informatics SIC Stuhlsatzenhausweg E1 4 Saarland Saarbrücken66123 Germany Trippstadter Str. 122 Rhineland-Palatinate Kaiserslautern67663 Germany

Monocular egocentric 3D human motion capture remains a significant challenge, particularly under conditions of low lighting and fast movements, which are common in head-mounted device applications. Existing methods that rely on RGB cameras often fail under these conditions. To address these limitations, we introduce EventEgo3D++, the first approach that leverages a monocular event camera with a fisheye lens for 3D human motion capture. Event cameras excel in high-speed scenarios and varying illumination due to their high temporal resolution, providing reliable cues for accurate 3D human motion capture. EventEgo3D++ leverages the LNES representation of event streams to enable precise 3D reconstructions. We have also developed a mobile head-mounted device (HMD) prototype equipped with an event camera, capturing a comprehensive dataset that includes real event observations from both controlled studio environments and in-the-wild settings, in addition to a synthetic dataset. Additionally, to provide a more holistic dataset, we include allocentric RGB streams that offer different perspectives of the HMD wearer, along with their corresponding SMPL body model. Our experiments demonstrate that EventEgo3D++ achieves superior 3D accuracy and robustness compared to existing solutions, even in challenging conditions. Moreover, our method supports real-time 3D pose updates at a rate of 140Hz. This work is an extension of the EventEgo3D approach (CVPR 2024) and further advances the state of the art in egocentric 3D human motion capture. For more details, visit the project page at https://***. Copyright © 2025, The Authors. All rights reserved.

关键词： Motion capture

Virtual reconstruction of the Red Room exhibition

学校读者我要写书评

暂无评论

Digital Applications in Archaeology and Cultural Heritage 2025年 37.0卷

作者： Guthe, Michael Mühlnikel, Marcus University of Bayreuth Visual Computing Universitätsstraße 30 Bavaria Bayreuth 95447 Germany University of Bayreuth Institute for Franconian Regional History Marktplatz 1 Bavaria Thurnau 95349 Germany

Museums increasingly use interactive installations to enhance exhibitions with additional content. Virtual reality can give the visitor access to remote places and bring the past back to life. This is especially interesting when artifacts cannot be brought back since they were sold or otherwise lost. In this project, we have partially reconstructed the Giech's family exhibition in the Red Room of Thurnau Castle. The exhibition ended when the family had to sell the exhibits in the 19th century because of financial problems. The goal of the project is to fully reconstruct the part of the exhibition that was shown in the Red Room based on a painting and the list of artifacts. While museums explain exhibits using information boards, the visitor of the Giech's family exhibition would have been guided by the earl himself. Therefore, we used virtual reality interaction techniques to make the exhibition more immersive and accessible. © 2025

关键词： Human–computer interfaces Virtual museum

Scaffolding Children's critical reflection on intelligent technologies: Opportunities from speculative fiction

学校读者我要写书评

暂无评论

International Journal of Child-Computer Interaction 2025年 43卷

作者： Malinverni, Laura Schaper, Marie-Monique Rubegni, Elisa Tamashiro, Mariana Aki Visual Arts and Design Department University of Barcelona Barcelona Spain Department of Digital Design and Information Studies Aarhus University Aarhus Denmark School of Computing and Communication Lancaster University Lancaster United Kingdom

Current technological development of Artificial Intelligence (AI) requires educational practices that address the social and ethical implications derived from these emerging technologies. To this end, an increasing number of educational practices are pursuing the goal of supporting children's critical reflection on these topics. Our research aims at understanding how speculative fiction-based resources can meet and respond to the goals of supporting children's critical reflection on AI technologies and their impact on society. Through revisiting relevant literature on these topics and critically analyzing our own practices in three different settings, we identify a set of opportunities and challenges oriented at guiding the design of resources capable of taking advantage of speculative fiction as a way to support critical reflection. © 2025 Elsevier B.V.

关键词： AI literacy Children Critical reflection Reflective AI literacy Speculative fiction