Depth Anything V4: Dynamic 4D Scene Reconstruction via Riemannian Flow Matching on 4D Gaussian Splatting
Depth Anything V4:基于4D高斯溅射的黎曼流匹配的动态4D场景重建
机构 * College of Artificial Intelligence, Nanjing University of Posts and Telecommunications(南京邮电大学人工智能学院) ; College of Automation, Nanjing University of Posts and Telecommunications(南京邮电大学自动化学院)
专题命中 Gaussian Splatting :Gaussian Splatting(title,abstract);分类 cs.CV
AI总结 本研究提出DAV4框架,将RFM应用于4D高斯溅射参数实现动态4D场景重建,其F分数较基线提升0.044,在相关任务上优于现有模型且无需人工标注深度标签。
Comments Major errors in research