构建了超千类3D物体数据集，助力生成与学习模型性能提升
UnCommon Objects in 3D
- 收集1000+类360°高分辨率视频，带3D标注与重建
- 相比已有数据集，3D模型在uCO3D上训练效果更优
- 适合3D生成、视觉建模及多视角学习研究者使用
我们提出Uncommon Objects in 3D（uCO3D），一个面向3D深度学习与3D生成AI的新型物体中心数据集。uCO3D是目前公开最大的高分辨率3D视频数据集，涵盖超过1000个物体类别，实现全360°覆盖。其多样性显著高于MVImgNet与CO3Dv2，且通过严格质量检查确保视频与3D标注的高质量。数据集包含3D相机位姿、深度图与稀疏点云标注，每个物体还配有文本描述和3D高斯溅射重建。我们在MVImgNet、CO3Dv2和uCO3D上训练多个大型3D模型，结果表明使用uCO3D训练的模型表现更优，证明其在学习任务中的优越性。
原文摘要 · Abstract (English)
We introduce Uncommon Objects in 3D (uCO3D), a new object-centric dataset for 3D deep learning and 3D generative AI. uCO3D is the largest publicly-available collection of high-resolution videos of objects with 3D annotations that ensures full-360$^{\circ}$ coverage. uCO3D is significantly more diverse than MVImgNet and CO3Dv2, covering more than 1,000 object categories. It is also of higher quality, due to extensive quality checks of both the collected videos and the 3D annotations. Similar to analogous datasets, uCO3D contains annotations for 3D camera poses, depth maps and sparse point clouds. In addition, each object is equipped with a caption and a 3D Gaussian Splat reconstruction. We train several large 3D models on MVImgNet, CO3Dv2, and uCO3D and obtain superior results using the latter, showing that uCO3D is better for learning applications.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。