3D-Fixer: Coarse-to-Fine In-place Completion for 3D Scenes from a Single Image
3D-Fixer:从单张图像生成3D场景的粗到细就地补全
机构 * College of Computer Science, Nankai University(南开大学计算机科学学院) ; School of Intelligence Science and Technology, Nanjing University(南京大学智能科学与技术学院) ; Horizon Robotics(地平线机器人) ; D-Robotics
AI总结 本文提出3D-Fixer,通过粗到细的生成方案解决单张图像生成3D场景的布局和3D资产恢复问题,结合双分支条件网络和Occlusion-Robust Feature Alignment策略,提升生成精度和效率,同时引入ARSG-110K数据集提升训练效果。
Comments 17 pages, 10 figures, CVPR 2026, project page: https://zx-yin.github.io/3dfixer