arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2103.13318cs.CV
收藏

不同表现领域和任务类型迁移学习的影响因素

Factors of Influence for Transfer Learning across Diverse Appearance Domains and Task Types

Google Research(谷歌研究院)

Thomas Mensink, Jasper Uijlings, Alina Kuznetsova, Michael Gygli, Vittorio Ferrari

摘要:

迁移学习能够重用在源任务上学到的知识来帮助学习目标任务。一种简单形式的迁移学习在当前最先进的计算机视觉模型中很常见,即在ILSVRC数据集上预训练用于图像分类的模型,然后对任何目标任务进行微调。然而,以前对迁移学习的系统研究是有限的,它预期工作的情况下没有完全理解。在本文中,我们对迁移学习进行了广泛的实验性探索,涉及不同的图像领域(消费者照片、自动驾驶、航拍图像、水下、室内场景、合成、特写)和任务类型(语义分割、目标检测、深度估计、关键点检测)。重要的是,这些都是与现代计算机视觉应用相关的复杂、结构化的输出任务类型。我们总共进行了2000多个迁移学习实验,其中包括许多源和目标来自不同图像域、任务类型或两者兼而有之的实验。我们系统地分析了这些实验,以了解图像域、任务类型和数据集大小对迁移学习性能的影响。我们的研究导致了几个见解和具体的建议:(1)对于大多数任务,存在一个显著优于ILSVRC'12预训练的源;(2)图像域是实现正迁移的最重要因素;(3)源数据集应该是目标数据集的图像域\emph{include}以实现最佳结果;(4)同时,当源任务的图像域比目标任务的图像域宽得多时,我们只观察到很小的负面影响;(5)跨任务类型的转移可能是有益的,但其成功在很大程度上取决于源和目标任务类型。

英文摘要:

Transfer learning enables to re-use knowledge learned on a source task to help learning a target task. A simple form of transfer learning is common in current state-of-the-art computer vision models, i.e. pre-training a model for image classification on the ILSVRC dataset, and then fine-tune on any target task. However, previous systematic studies of transfer learning have been limited and the circumstances in which it is expected to work are not fully understood. In this paper we carry out an extensive experimental exploration of transfer learning across vastly different image domains (consumer photos, autonomous driving, aerial imagery, underwater, indoor scenes, synthetic, close-ups) and task types (semantic segmentation, object detection, depth estimation, keypoint detection). Importantly, these are all complex, structured output tasks types relevant to modern computer vision applications. In total we carry out over 2000 transfer learning experiments, including many where the source and target come from different image domains, task types, or both. We systematically analyze these experiments to understand the impact of image domain, task type, and dataset size on transfer learning performance. Our study leads to several insights and concrete recommendations: (1) for most tasks there exists a source which significantly outperforms ILSVRC'12 pre-training; (2) the image domain is the most important factor for achieving positive transfer; (3) the source dataset should \emph{include} the image domain of the target dataset to achieve best results; (4) at the same time, we observe only small negative effects when the image domain of the source task is much broader than that of the target; (5) transfer across task types can be beneficial, but its success is heavily dependent on both the source and target task types.

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。

↑