A Camera-Native Stereo VR180 Dataset

Computer Science > Computer Vision and Pattern Recognition arXiv:2610.10607 (cs) [Submitted on 7 Oct 2026] Title: A Camera-Native Stereo VR180 Dataset Authors: Linxuan Lu.

计算机科学 > 计算机视觉与模式识别 arXiv:2610.10607 (cs) [提交于 2026 年 10 月 7 日] 标题：一个相机原生的立体 VR180 数据集 作者：Linxuan Lu。

Abstract: Immersive VR180 video is increasingly produced with professional stereo fisheye cameras, yet public VR180 research resources are mostly collected from online platforms such as YouTube: already stitched, projected and compressed by unknown pipelines, and without lens calibration.

摘要：沉浸式 VR180 视频越来越多地由专业立体鱼眼相机制作，但公开的 VR180 研究资源大多是从 YouTube 等在线平台收集的：这些视频已经过拼接、投影和压缩，且处理流程未知，也缺乏镜头校准。

We present a firsthand-captured stereo VR180 dataset recorded with two Blackmagic URSA Cine Immersive cameras. It contains 1,211 samples — 636 stereo video clips (2,220.8 s, mostly 90 fps) and 575 stereo stills — each released as camera-native Blackmagic RAW, separate-eye native fisheye HEVC (8160x7200 per eye) and half-equirectangular HEVC (7200x7200 per eye), together with the factory lens calibration, portable fisheye/half-equirectangular conversion tools and AI-generated scene and visual-challenge annotations.

我们展示了一个使用两台 Blackmagic URSA Cine Immersive 相机录制的一手立体 VR180 数据集。它包含 1,211 个样本——636 个立体视频片段（2,220.8 秒，大多为 90 fps）和 575 张立体静态图像——每个样本均以相机原生的 Blackmagic RAW、分眼原生鱼眼 HEVC（每眼 8160x7200）和半等距柱状投影 HEVC（每眼 7200x7200）格式发布，并附带原厂镜头校准、便携式鱼眼/半等距柱状投影转换工具以及人工智能生成的场景和视觉挑战标注。

Re-encoding the released fisheye and half-equirectangular renders with x265 over 24 clips, both eyes, four rate points and nine viewing directions, native-fisheye coding needed more bitrate than half-equirectangular coding at equal viewport quality for all 24 clips (median +38%), in every part of the field of view.

通过使用 x265 对发布的鱼眼和半等距柱状投影渲染图进行重新编码（涵盖 24 个片段、双眼、四个速率点和九个观看方向），结果显示在所有 24 个片段中，在相同的视口质量下，原生鱼眼编码比半等距柱状投影编码需要更高的比特率（中位数增加 38%），且覆盖视野的每个部分。