新方法无需旧数据也能避免模型遗忘，提升高光谱图像增量分类精度。
HSI Image Enhancement Classification Based on Knowledge Distillation: A Study on Forgetting
- 用新类别样本传递旧类别知识，不依赖历史数据
- 通过掩码筛选关键知识，减少错误信息干扰
- 适合持续学习场景，尤其对数据存储受限的任务
在高光谱图像的增量分类任务中，灾难性遗忘是不可避免的挑战。尽管记忆回放方法可缓解此问题，但严重依赖旧类别的样本。本文提出一种基于教师的知识保留方法，利用增量类别样本来缓解模型对旧类别样本的遗忘，无需依赖旧类别样本。此外，引入一种基于掩码的部分类别知识蒸馏算法，通过解耦知识蒸馏过程，过滤掉可能误导学生模型的错误信息，从而提升整体分类准确率。对比实验与消融实验验证了该方法的鲁棒性能。
原文摘要 · Abstract (English)
In incremental classification tasks for hyperspectral images, catastrophic forgetting is an unavoidable challenge. While memory recall methods can mitigate this issue, they heavily rely on samples from old categories. This paper proposes a teacher-based knowledge retention method for incremental image classification. It alleviates model forgetting of old category samples by utilizing incremental category samples, without depending on old category samples. Additionally, this paper introduces a mask-based partial category knowledge distillation algorithm. By decoupling knowledge distillation, this approach filters out potentially misleading information that could misguide the student model, thereby enhancing overall accuracy. Comparative and ablation experiments demonstrate the proposed method's robust performance.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。