will explore how advances in generative networks and deep-learning approaches to inpainting can be combined to create virtual camera views

将探索如何结合生成网络和深度学习修复方法的进步来创建虚拟相机视图

• 批准号: 2288914
• 金额: --
• 依托单位: University of Surrey
• 依托单位国家: 英国
• 项目类别: Studentship
• 财政年份: 2019
• 资助国家: 英国
• 起止时间: 2019 至 无数据
• 项目状态: 已结题
• 来源: https://gtr.ukri.org/projects?ref=studentship-2288914
• 关键词: will; explore; how; advances; generative

Generating virtual camera views using a Generative Query Network This project will explore how advances in Generative networks and deep-learning approaches to inpainting can be combined to create virtual camera views. A unifiedgenerative deep learning framework for viewpoint interpolation and inpainting, should effectively blend the observed viewpoints and hallucinated information, to produce a virtual camera feed from otherwise impractical vantage points. Thedeveloped technology is of interest as a potential extension to the capabilities of the Ed system currently being developed by BBC R&D (using AI techniques for automated content editing).

该项目将探索如何将生成网络和深度学习方法相结合，以创建虚拟相机视图。一个用于视点插值和绘制的统一生成深度学习框架，应该有效地混合观察到的视点和幻觉信息，从其他不切实际的有利位置产生虚拟摄像机馈送。开发的技术是BBC研发部门目前正在开发的Ed系统（使用人工智能技术进行自动内容编辑）功能的潜在扩展。

项目成果

SaiNet: Stereo aware inpainting behind objects with generative networks

• DOI: 10.48550/arxiv.2205.07014
• 发表时间: 2022-05
• 期刊: ArXiv
• 作者: Violeta Men'endez Gonz'alez;Andrew Gilbert;Graeme Phillipson;Stephen Jolly;Simon Hadfield

SVS: Adversarial refinement for sparse novel view synthesis

SVS：稀疏新颖视图合成的对抗性细化

• 发表时间: 2022
• 作者: Menendez Gonzalez V