arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2104.02290cs.CV
收藏

对比同实推广

Contrastive Syn-to-Real Generalization

The University of Texas at Austin(德克萨斯大学奥斯汀分校);NVIDIA(英伟达);California Institute of Technology(加州理工学院)

Wuyang Chen, Zhiding Yu, Shalini De Mello, Sifei Liu, Jose M. Alvarez, Zhangyang Wang, Anima Anandkumar

摘要:

在合成数据上进行培训对于标签或数据稀缺的情况是有益的。然而,综合训练的模型往往遭受在真实领域由于领域差距的泛化能力差。在这项工作中,我们做了一个关键的观察,学习的特征嵌入的多样性在泛化性能中起着重要的作用。为此,我们提出了对比合成到真实泛化(CSG),这是一种新的框架,它利用预先训练的ImageNet知识来防止合成域的过拟合,同时促进特征嵌入的多样性作为归纳偏差来提高泛化能力。此外,我们加强了建议CSG框架与注意力池(A-pool),让模型专注于语义上的重要区域,并进一步提高其泛化。我们证明了CSG在各种综合训练任务上的有效性,在zero-shot领域泛化上表现出最先进的性能。

英文摘要:

Training on synthetic data can be beneficial for label or data-scarce scenarios. However, synthetically trained models often suffer from poor generalization in real domains due to domain gaps. In this work, we make a key observation that the diversity of the learned feature embeddings plays an important role in the generalization performance. To this end, we propose contrastive synthetic-to-real generalization (CSG), a novel framework that leverages the pre-trained ImageNet knowledge to prevent overfitting to the synthetic domain, while promoting the diversity of feature embeddings as an inductive bias to improve generalization. In addition, we enhance the proposed CSG framework with attentional pooling (A-pool) to let the model focus on semantically important regions and further improve its generalization. We demonstrate the effectiveness of CSG on various synthetic training tasks, exhibiting state-of-the-art performance on zero-shot domain generalization.

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。

↑