从自我中心物体操作中自监督视觉学习的计算解释
A Computational Account Of Self-Supervised Visual Learning From Egocentric Object Play
AI总结：
本文基于 Toybox 自我中心视频数据集，研究将同一物体不同视角表征等同的自监督对比学习方法，证明其能鲁棒提升下游图像分类性能并可迁移到多个任务。
摘要：
儿童发展研究表明，操作物理物体的具身经验有助于多种认知能力，包括视觉学习。这种经验的一个特点是学习者会从多个不同视角观察同一物体。本文研究将不同视角等同起来的学习信号——例如为同一物体的不同视图分配相似表征——如何支持鲁棒的视觉学习。我们使用 Toybox 数据集，该数据集包含人类操作不同物体的自我中心视频，并利用自监督对比学习的计算机视觉框架进行实验。我们发现，通过等同物体不同物理视角所学到的表征有助于提升下游图像分类准确率。进一步实验表明，这种性能提升对视角间差距的变化具有鲁棒性，并且其收益可迁移到多个不同的图像分类任务。
英文摘要：
Research in child development has shown that embodied experience handling physical objects contributes to many cognitive abilities, including visual learning. One characteristic of such experience is that the learner sees the same object from several different viewpoints. In this paper, we study how learning signals that equate different viewpoints -- e.g., assigning similar representations to different views of a single object -- can support robust visual learning. We use the Toybox dataset, which contains egocentric videos of humans manipulating different objects, and conduct experiments using a computer vision framework for self-supervised contrastive learning. We find that representations learned by equating different physical viewpoints of an object benefit downstream image classification accuracy. Further experiments show that this performance improvement is robust to variations in the gaps between viewpoints, and that the benefits transfer to several different image classification tasks.
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。
