arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1802.02568cs.CVcs.LG
收藏

VISER:视觉自正则化

VISER: Visual Self-Regularization

University of Washington(华盛顿大学);Flickr, Yahoo Research(雅虎研究公司Flickr)

Hamid Izadinia, Pierre Garrigues

摘要:

在这项工作中,我们提出了使用大量的未标记图像作为正则化数据的来源,学习强大的视觉表示。给定一个由标记数据集以监督方式训练的视觉模型,我们通过合并大量未标记数据来增加我们的训练样本,并训练半监督模型。我们证明了我们提出的学习方法利用了大量的未标记图像,并提高了视觉识别性能,这表明需要依赖大型标记数据集来学习鲁棒的表示。在我们的方法中,为了增加学习鲁棒视觉模型所需的图像实例的数量,每个标记的图像将其标签传播到最近的未标记图像实例。这些检索到的未标记图像作为每个标记图像的局部扰动,以执行视觉自正则化(VISER)。为了检索这种视觉自正则化器,我们计算由全卷积神经网络中的倒数第二层定义的语义空间中的余弦相似性。我们使用公开的Yahoo Flickr Creative Commons 100M数据集作为我们的未标记图像集的源,并提出了一个分布式的近似最近邻算法,使检索实用的规模。使用标记的实例及其正则化样本,我们表明,我们显着提高对象的MS COCO和Visual Genome数据集上的对象出现在上下文中的分类和本地化性能。

英文摘要:

In this work, we propose the use of large set of unlabeled images as a source of regularization data for learning robust visual representation. Given a visual model trained by a labeled dataset in a supervised fashion, we augment our training samples by incorporating large number of unlabeled data and train a semi-supervised model. We demonstrate that our proposed learning approach leverages an abundance of unlabeled images and boosts the visual recognition performance which alleviates the need to rely on large labeled datasets for learning robust representation. To increment the number of image instances needed to learn robust visual models in our approach, each labeled image propagates its label to its nearest unlabeled image instances. These retrieved unlabeled images serve as local perturbations of each labeled image to perform Visual Self-Regularization (VISER). To retrieve such visual self regularizers, we compute the cosine similarity in a semantic space defined by the penultimate layer in a fully convolutional neural network. We use the publicly available Yahoo Flickr Creative Commons 100M dataset as the source of our unlabeled image set and propose a distributed approximate nearest neighbor algorithm to make retrieval practical at that scale. Using the labeled instances and their regularizer samples we show that we significantly improve object categorization and localization performance on the MS COCO and Visual Genome datasets where objects appear in context.

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。

↑