ACM Multimedia 2025

International Workshop on Application-driven Point Cloud Processing and 3D Vision (APP3DV)


Conference Venue: Dublin, Ireland

Conference Date: October 27 to 31, 2025



Overview

Point cloud is an explicit representation for 3D objects and scenes, providing precise coordinates and attributes to depict the real world. It lies a solid foundation for wide applications, such as digital entertainment, autonomous driving, remote sensing, robotics, unmanned aerial vehicles, etc. The huge data amount of point clouds motivates the research and development of efficient compression algorithms to relieve the burden of data transmission and storage. Therefore, international industry companies, research institutions and standardization organizations become very interested in devising better compression methods, as well as powerful standards. Moreover, point cloud enhancement algorithms are also very popular, including upsampling, denoising, completion, compression artifacts removal, frame interpolation, etc., which play a critical role in elevating the point cloud quality. Besides the quality evaluation and enhancement for human visual experience, researchers also pay much attention on those for machine analysis tasks. Due to increased efforts on immersive media, embodied artificial intelligence (AI) and unmanned system applications, the 3D vision analysis based on point clouds or point cloud-based multi-modalities is becoming a critical research trend for artificial intelligence. Point cloud technologies can boost the advancements of both the traditional analysis tasks, including classification, segmentation, detection, tracking, etc., and the emerging analysis tasks, including captioning, grounding, task decomposition, question-answering, navigation, etc. The 3D reconstruction and rendering tasks of point clouds have promoted the emergence of 3D Gaussian splatting, which have raised new explorations of 3D visual data compression, reconstruction, generation, rendering, and representation. The advanced algorithms on point cloud compression, enhancement and analysis are solicited, and contributions that advance the state-of-the-art 3D Gaussian splatting technologies for 3D visual data compression, reconstruction, generation, rendering, and representation are also welcomed.

Call for Papers

This workshop intends to provide a forum for researchers and engineers to present their latest innovations and share their experiences on all aspects in the field of point cloud processing and 3D vision, as well as the advancements in the efforts for standardization, system implementation and emerging applications. Topics of interest include, but are not limited to:
  • Datasets and modeling for human and machine perceptions of 3D visual data
  • Point cloud compression and transmission (deep-learning-based and non-deep-learning-based)
  • Point cloud enhancement (upsampling, denoising, completion, compression artifacts removal, frame interpolation)
  • Point cloud analysis (classification, segmentation, detection, tracking)
  • Gaussian splatting for 3D visual data compression
  • Gaussian splatting for 3D reconstruction, generation, and 6DOF rendering
  • Gaussian splatting for 3D representation and understanding
  • Multimodal learning for captioning, grounding, task decomposition, question-answering, navigation
  • Foundation models and generative models for 3D vision and multimodal learning
  • System implementations and standards of point cloud, Gaussian splatting and multimodal technologies
  • Emerging point cloud, Gaussian splatting and multimodal technologies in immersive media, autonomous driving, embodied AI, and unmanned systems
Important Dates:
  • Call for Papers: 2025/3/15
  • Paper Submission Deadline (Including Fast-Track Submissions for Rejected Papers at ACM MM 2025): 2025/7/11 (23:59 GMT)
  • Paper Notification: 2025/8/1
  • Paper Camera-ready: 2025/8/11
  • ACM MM 2025 Workshop Date: 2025/10/27-28 (Dublin, Ireland)

Paper Submission

Format: Submitted papers (.pdf format) must use the ACM Article Template https://www.acm.org/publications/proceedings-template. Please remember to add Concepts and Keywords.
Length: Submissions can be of varying length from 4 to 8 pages, plus additional pages for the reference pages; i.e., the reference page(s) are not counted to the page limit of 4 to 8 pages. There is no distinction between long and short papers, but the authors may themselves decide on the appropriate length of the paper. All papers will undergo the same review process and review period.
Reviewing Process: Paper submissions must conform with the “double-blind” review policy. All papers will be peer-reviewed by experts in the field, they will receive at least three reviews. Acceptance will be based on relevance to the workshop, scientific novelty, and technical quality. The workshop papers will be published in the ACM Digital Library.
Submission Site: https://openreview.net/group?id=acmmm.org/ACMMM/2025/Workshop/APP3DV.

Workshop Schedule

The APP3DV workshop has been scheduled to 13:30 to 17:00 (with a break from 15:00 to 15:30) in October 28, 2025, and the conference venue is Radisson-Swift at ACM Multimedia 2025 Conference.

Session 1: 3D Point Cloud Compression and Processing 13:30 to 15:00, October 28, 2025
Authors Paper Title
Yuxin Zhan (School of Electronic and Computer Engineering, Peking University Shenzhen Graduate School), Tai Qin (School of Electronic and Computer Engineering, Peking University Shenzhen Graduate School), Shiqiang Long (Guangdong Bohua UHD Innovation Center Co., Ltd.), Weimin Zhang (Guangdong Bohua UHD Innovation Center Co., Ltd.), Ge Li (School of Electronic and Computer Engineering, Peking University Shenzhen Graduate School) Slice-based Lossless Point Cloud Attribute Compression Using Intra Block Prediction
Tianfang Du (Fuzhou University), Jingjing Xu (Fuzhou University), Jian Chen (Fuzhou University), Mingkui Zheng (Fuzhou University), Tiesong Zhao (Fuzhou University Joint Semantic and Temporal Compression of LiDAR Point Clouds for Intelligent Driving
Monica Vahdati (MCRLab, School of Electrical Engineering and Computer Science, University of Ottawa), Kamran Gholizadeh HamlAbadi (MCRLab, School of Electrical Engineering and Computer Science, University of Ottawa), Haiwei Dong (Ottawa Research Center, Huawei), Basim Hafidh (MCRLab, School of Electrical Engineering and Computer Science, University of Ottawa), Abdulmotaleb El Saddik (MCRLab, School of Electrical Engineering and Computer Science, University of Ottawa) Real-Time 2D-to-3D LiDAR Reconstruction and Omniverse-Based Visualization for Scalable Point Cloud Applications
Mengting Liu (Beijing University of Chemical Technology), Haijiang Zhu (Beijing University of Chemical Technology), Ning An (China Coal Research Institute) U4SNet: Dynamic Class Balanced Hierarchical 3D Underground Semi-Supervised Semantic Segmentation
Milo Beliën (Information and Computing Sciences, Utrecht University (ICS)), Metehan Doyran (Information and Computing Sciences, Utrecht University) Two-Step Deep Learning Approach for Fine-Grained Railway Catenary Segmentation
Break 15:00 to 15:30, October 28, 2025
Session 2: 3D Gaussian Splatting Compression and Processing 15:30 to 17:00, October 28, 2025
Authors Paper Title
Weitao Xiong (The Hong Kong University of Science and Technology and Xiamen University Malaysia), Guo Zhu (The Chinese University of Hong Kong, Shenzhen), Kunyuan Cai (The Chinese University of Hong Kong, Shenzhen and Xiamen University Malaysia) Implicit Neural Representations for 3D Gaussian Compression
Takuya Fujihashi (The University of Osaka), Akihiro Kuwabara (The University of Osaka), Toshiaki Koike-Akino (Mitsubishi Electric Research Laboratories) QKAN-GS: Quantum-Empowered 3D Gaussian Splatting
Wei Sun (University of the Chinese Academy of Sciences), Yuan Li (University of the Chinese Academy of Sciences), Jianbin Jiao (University of the Chinese Academy of Sciences) MaskGaussians: Advancing Open-Vocabulary 3D Scene Understanding with Dynamic Masking
Hyunseo Koh (Soongsil University), Seunghyun Oh (Soongsil University), Jungyun Jang (Soongsil University), Heewon Kim (Soongsil University) Intrinsic-Guided Photorealistic Style Transfer for Radiance Fields

History and References

[1] Wei Gao, Sam Kwong, Zhu Li, Shan Liu, Ge Li, “APP3DV'25: International Workshop on Application-driven Point Cloud Processing and 3D Vision,” Proceedings of the 33rd ACM International Conference on Multimedia, 2025.
BibTeX
    @inproceedings{gao2025app3dv, title={APP3DV'25: International Workshop on Application-driven Point Cloud Processing and 3D Vision}, author={Gao, Wei and Kwong, Sam and Li, Zhu and Liu, Shan and Li, Ge}, booktitle={Proceedings of the 33rd ACM International Conference on Multimedia}, year={2025} }
[2] Wei Gao, Ge Li, Hui Yuan, Raouf Hamzaoui, Zhu Li, Shan Liu, “APCCPA’22: 1st International Workshop on Advances in Point Cloud Compression, Processing and Analysis,” Proceedings of the 30th ACM International Conference on Multimedia, 2022.
BibTeX
    @inproceedings{gao2022apccpa, title={APCCPA'22: 1st International Workshop on Advances in Point Cloud Compression, Processing and Analysis}, author={Gao, Wei and Li, Ge and Yuan, Hui and Hamzaoui, Raouf and Li, Zhu and Liu, Shan}, booktitle={Proceedings of the 30th ACM International Conference on Multimedia}, pages={7392--7393}, year={2022} }
[3] Wei Gao, Ge Li, “Point Cloud Compression, Enhancement and Applications: From 3D Perception to Large Models,” Proceedings of the 32nd ACM International Conference on Multimedia, 2024.
BibTeX
    @inproceedings{gao2024point, title={Point Cloud Compression, Enhancement and Applications: From 3D Perception to Large Models}, author={Gao, Wei and Li, Ge}, booktitle={Proceedings of the 32nd ACM International Conference on Multimedia}, pages={11292--11293}, year={2024} }
[4] Wei Gao, Ge Li, “Deep Learning for 3D Point Clouds,” Springer, 2025.
BibTeX
    @book{gao2025deep, title={Deep Learning for 3D Point Clouds}, author={Gao, Wei and Li, Ge}, year={2025}, publisher={Springer} }
[5] Ge Li, Wei Gao, Wen Gao, “Point Cloud Compression: Technologies and Standardization,” Springer, 2024.
BibTeX
    @book{li2024point, title={Point Cloud Compression: Technologies and Standardization}, author={Li, Ge and Gao, Wei and Gao, Wen}, year={2024}, publisher={Springer Nature} }

Organizers

Wei Gao

Peking University, China
Sam Kwong

Lingnan University, Hong Kong
Zhu Li

University of Missouri-Kansas City, USA
Shan Liu

Tencent Media Lab, USA
Ge Li

Peking University, China