研究人员开发了一个新的可微分框架，使用结构光和单个摄像头同时捕获物体的形状和反射率。该方法自适应地计算照明条件，以减少深度不确定性，从而改进深度图和反射率参数重建。此外，还为个性化联邦学习引入了一个名为 VPDR 的新隐私插件，该插件通过自适应地分配噪声以保持语义可分离性来增强隐私-实用性权衡。 AI
影响 这些论文探讨了计算机视觉在物体重建和隐私保护联邦学习方面的新技术，可能推动这些领域的研究。
排序理由 该集群包含两篇不同的 arXiv 论文。
AI 生成摘要 · Google Gemini · 来自 4 个来源。 我们如何撰写摘要 →
arXiv:2605.06214v1 Announce Type: new Abstract: We present a differentiable framework to adaptively compute 4D illumination conditions with respect to an object, for efficient, high-quality simultaneous acquisition of its shape and reflectance, with a unified spatial-angular stru…
We present a differentiable framework to adaptively compute 4D illumination conditions with respect to an object, for efficient, high-quality simultaneous acquisition of its shape and reflectance, with a unified spatial-angular structured light and a single camera. Using a simple…
arXiv:2604.27833v1 Announce Type: new Abstract: Prototype-based Personalized Federated Learning (ProtoPFL) enables efficient multi-domain adaptation by communicating compact class prototypes, but directly sharing them poses privacy risks. A common defense involves per-example $\e…
Prototype-based Personalized Federated Learning (ProtoPFL) enables efficient multi-domain adaptation by communicating compact class prototypes, but directly sharing them poses privacy risks. A common defense involves per-example $\ell_2$ clipping before prototype computation to b…