FeDepth: Federated Learning for Depth Estimation under Robot Heterogeneity
FeDepth:面向机器人异构性的深度估计联邦学习
机构 * Ulsan National Institute of Science and Technology (UNIST)(蔚山科学技术院)
专题命中 机器人数据与评测 :robotic(abstract);分类 cs.RO、cs.CV
AI总结 针对机器人感知深度估计中联邦学习因客户端异构性导致性能下降的问题,本文提出FeDepth框架,通过软聚类建模客户端关系,在多架构上提升了联邦学习的鲁棒性。
Comments Accepted at ECCV 2026. Ganghyeon Lee and Inha Lee contributed equally. Kyungdon Joo is the corresponding author