通过显著性引导增强提升行为克隆中的视觉域鲁棒性
Enhancing Visual Domain Robustness in Behaviour Cloning via Saliency-Guided Augmentation
浏览论文内容
中文总结 AI 辅助
该研究针对行为克隆中常规增强方法应对视觉域偏移效果差的问题，提出RoboSaGA显著性引导增强方法，提升了模型的视觉域鲁棒性且保留了域内性能。
中文摘要 AI 辅助
在基于视觉的行为克隆（BC）中，随机裁剪、颜色抖动等常规图像增强方法在存在阴影、干扰项、背景变化等大幅视觉域偏移时往往效果不佳。叠加式增强方法通过混合域内和域外图像，在计算机视觉中展现出提升泛化能力的潜力，但其是否适用于BC仍不确定，因为必须保留任务关键语义、时空关系以及智能体-目标交互。为解决该问题，我们提出RoboSaGA，一种叠加式家族内的显著性引导增强方法，专为基于视觉的BC定制。RoboSaGA利用策略驱动的显著性在像素级别动态调整增强强度，在任务无关区域进行激进增强，同时保留任务关键信息。它可无缝集成到现有架构中，无需结构修改或额外学习目标。在仿真和真实场景中的实验表明，RoboSaGA在保留域内性能的同时，大幅提升了对视觉域偏移的鲁棒性，包括干扰项、背景变化以及光照和阴影变化。代码可在该https URL获取。
英文摘要
In vision-based behavior cloning (BC), conventional image augmentations such as Random Crop and Color Jitter often fall short under substantial visual domain shifts, including changes in shadows, distractors, and backgrounds. Superimposition-based augmentations, which blend in-domain and out-of-domain images, have shown promise for improving generalization in computer vision, but their suitability for BC remains uncertain because task-critical semantics, spatiotemporal relationships, and agent-target interactions must be preserved. To address this, we introduce RoboSaGA, a Saliency-Guided Augmentation method within the superimposition family tailored for vision-based BC. RoboSaGA dynamically adjusts augmentation intensity at the pixel level using policy-driven saliency, enabling aggressive augmentation in task-irrelevant regions while preserving task-critical information. It integrates seamlessly into existing architectures without requiring structural modifications or additional learning objectives. Experiments in both simulated and real-world settings show that RoboSaGA preserves in-domain performance while substantially improving robustness to visual domain shifts, including distractor and background changes, as well as lighting and shadow variations. Code is available at https://github.com/Zheyu-Zhuang/RoboSaGA.
发表机构
- KTH Royal Institute of Technology（皇家理工学院）
- Aalto University（阿尔托大学）
机构由 AI 辅助整理，请以论文原文为准。
