Hyper 360—Towards a Unified Tool Set Supporting Next Generation VR Film and TV Productions

• 作者: Antonis Karakottas Barnabas Takacs Dimitrios Zarpalas Hannes Fassold Nikolaos Zioulis Petros Daras Zsuzsanna Vincze
• 来源期刊: 软件工程与应用(英文)
• 学科: 工学
• 关键词: Deep Learning TENSOR Flow YoloV3 360? VIDEO Virtual REALITY Free VIEWPOINT VIDEO Quality Control Automatic CINEMATOGRAPHY MULTI-CAMERA Systems
• 年，卷（期）: 2019,（5）
• 页码范围: 127-148
• 页数: 22页
• 分类号: TP39

摘要

We describe four fundamental challenges that complex real-life Virtual Reality (VR) productions are facing today (such as multi-camera management, quality control, automatic annotation with cinematography and 360˚?depth estimation) and describe an integrated solution, called Hyper 360, to address them. We demonstrate our solution and its evaluation in the context of practical productions and present related results.