Stem-OB：基于类干收敛观测与扩散反演的可泛化视觉模仿学习
Stem-OB: Generalizable Visual Imitation Learning with Stem-Like Convergent Observation through Diffusion Inversion
AI总结：
针对视觉模仿学习在光照和纹理等扰动下泛化性差的问题，提出Stem-OB方法，利用预训练图像扩散模型进行图像反演以抑制低级视觉差异并保持场景结构，作为即插即用方案显著提升了真实世界任务的成功率。
摘要：
视觉模仿学习方法展现出强大的性能，但在面临光照和纹理等视觉输入扰动时缺乏泛化能力，阻碍了其实际应用。我们提出了Stem-OB，利用预训练的图像扩散模型来抑制低级视觉差异，同时保持高级场景结构。这一图像反演过程类似于将观测转换为一种共享表示，其他观测均由此衍生，同时去除了无关细节。Stem-OB与数据增强方法形成对比，因为它对各种未指定的外观变化具有鲁棒性，且无需额外训练。我们的方法是一种简单而高效的即插即用解决方案。实验结果证实了该方法在模拟任务中的有效性，并显示出在真实世界应用中的显著提升，与最佳基线相比，成功率平均提高了22.2%。详情请访问https://hukz18.github.io/Stem-Ob/。
英文摘要：
Visual imitation learning methods demonstrate strong performance, yet they lack generalization when faced with visual input perturbations, including variations in lighting and textures, impeding their real-world application. We propose Stem-OB that utilizes pretrained image diffusion models to suppress low-level visual differences while maintaining high-level scene structures. This image inversion process is akin to transforming the observation into a shared representation, from which other observations stem, with extraneous details removed. Stem-OB contrasts with data-augmentation approaches as it is robust to various unspecified appearance changes without the need for additional training. Our method is a simple yet highly effective plug-and-play solution. Empirical results confirm the effectiveness of our approach in simulated tasks and show an exceptionally significant improvement in real-world applications, with an average increase of 22.2% in success rates compared to the best baseline. See https://hukz18.github.io/Stem-Ob/ for more info.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
