论文概要
研究领域: CV 作者: Jiekai Wu, Rong Fu, Chuangqi Li 等 发布时间: 2026-04-03 arXiv: 2604.03212
中文摘要
真实部署中的遥感分割本质上是持续性的：新的语义类别不断出现，采集条件随季节、城市和传感器而变化。我们提出ProtoFlow，一个时间感知原型动态框架，将类原型建模为轨迹，并通过显式时间向量场学习其演化。在标准类和域增量遥感基准上的实验显示相比强基线有持续增益，mIoUall提升1.5-2.0点，同时减少遗忘。
原文摘要
Remote sensing segmentation in real deployment is inherently continual: new semantic categories emerge, and acquisition conditions shift across seasons, cities, and sensors. Despite recent progress, many incremental approaches still treat training steps as isolated updates, which leaves representation drift and forgetting insufficiently controlled. We present ProtoFlow, a time-aware prototype dynamics framework that models class prototypes as trajectories and learns their evolution with an explicit temporal vector field. By jointly enforcing low-curvature motion and inter-class separation, ProtoFlow stabilizes prototype geometry throughout incremental learning. Experiments on standard class- and domain-incremental remote sensing benchmarks show consistent gains over strong baselines, inclu...
--- *自动采集于 2026-04-06*
#论文 #arXiv #CV #小凯