3D Modeling, Animation and Perception guided Compression for Videoconferencing

用于视频会议的 3D 建模、动画和感知引导压缩

• 批准号: 548959-2019
• 负责人: Basu, Anup
• 金额: $ 2.19万
• 依托单位: University of Alberta
• 依托单位国家: 加拿大
• 项目类别: Alliance Grants
• 财政年份: 2021
• 资助国家: 加拿大
• 起止时间: 2021-01-01 至 2022-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=722524
• 关键词: 3D; Modeling; Animation; Perception; guided

The goal of this partnership is to develop the ability to transmit multimedia visual information to mobile devices using very low bandwidth while maintaining high perceptual quality. The outcome will be reducing bandwidth requirements for everyday users, thereby reducing costs and increasing access. Video alone uses up nearly 90% of the bandwidth over the Internet. With even more bandwidth used considering images and graphics. Thus, reducing the bandwidth requirements for video will have a very big impact on the cost and performance associated with Internet use for both network providers and end users. Unlike conventional video compression this project will investigate first extracting distinct objects from an image and then using 3D graphical models and animation to recreate a realistic representation of a real dynamic scene. The project topic is a critical step towards introducing the next generation of video representation and delivery to Canadians. The outcomes of the project will enable much higher quality videoconferencing, compared to the fuzzy video with freeze frames that we routinely encounter.

这一合作伙伴关系的目标是开发使用极低带宽向移动设备传输多媒体视觉信息的能力，同时保持高感知质量。其结果将是降低日常用户的带宽需求，从而降低成本并增加访问。仅视频一项就占用了互联网近90%的带宽。考虑到图像和图形，使用的带宽甚至更多。因此，减少视频的带宽需求将对网络提供商和最终用户使用互联网的成本和性能产生非常大的影响。与传统的视频压缩不同，该项目将首先研究从图像中提取不同的对象，然后使用3D图形模型和动画来重建真实动态场景的逼真表示。该项目的主题是向加拿大人介绍下一代视频表现和交付的关键一步。与我们经常遇到的带有定格的模糊视频相比，该项目的成果将使高质量的视频会议成为可能。