PulseAugur
LIVE 15:06:37
research · [4 sources] ·
0
research

New research explores adaptive 4D illumination for shape/reflectance capture and privacy-preserving federated…

By PulseAugur Editorial · Summary by gemini-2.5-flash-lite from 4 sources

Researchers have developed a new differentiable framework for capturing an object's shape and reflectance simultaneously using structured light and a single camera. This method adaptively computes illumination conditions to reduce depth uncertainty, leading to improved depth map and reflectance parameter reconstructions. Separately, a novel privacy plug-in called VPDR has been introduced for personalized federated learning, which enhances privacy-utility trade-offs by adaptively allocating noise to preserve semantic separability. AI

Summary written by gemini-2.5-flash-lite from 4 sources. How we write summaries →

IMPACT These papers explore novel techniques in computer vision for object reconstruction and privacy-preserving federated learning, potentially advancing research in these areas.

RANK_REASON The cluster contains two distinct academic papers submitted to arXiv.

Read on arXiv cs.CV →

COVERAGE [4]

  1. arXiv cs.CV TIER_1 · Huakeng Ding, Yaowen Chen, Kun Zhou, Hongzhi Wu ·

    Differentiable Adaptive 4D Structured Illumination for Joint Capture of Shape and Reflectance

    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…

  2. arXiv cs.CV TIER_1 · Hongzhi Wu ·

    Differentiable Adaptive 4D Structured Illumination for Joint Capture of Shape and Reflectance

    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…

  3. arXiv cs.CV TIER_1 · Yuhua Wang, Qinnan Zhang, Xiaodong Li, Huan Zhang, Yifan Sun, Wangjie Qiu, Hainan Zhang, Yongxin Tong, Zhiming Zheng ·

    Taming Noise-Induced Prototype Degradation for Privacy-Preserving Personalized Federated Fine-Tuning

    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…

  4. arXiv cs.CV TIER_1 · Zhiming Zheng ·

    Taming Noise-Induced Prototype Degradation for Privacy-Preserving Personalized Federated Fine-Tuning

    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…

RELATED ENTITIES

RELATED TOPICS