arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1704.01285cs.CV
收藏

深度度量学习的智能挖掘

Smart Mining for Deep Metric Learning

The University of Adelaide(阿德莱德大学);Monash University(莫纳什大学)

Ben Harwood, Vijay Kumar B G, Gustavo Carneiro, Ian Reid, Tom Drummond

摘要:

为了解决深度度量学习问题和生成特征嵌入,当前的方法通常使用三元组模型来最小化来自同一类的样本之间的相对距离,并最大化来自不同类的样本之间的相对距离。虽然成功了,但这个三元组模型的训练收敛性可能会受到绝大多数训练样本将产生幅度接近于零的梯度的事实的影响。这个问题激发了探索嵌入的全局结构的方法和探索硬负/正挖掘的其他方法的发展。这种采矿方法的有效性往往与棘手的计算要求有关。在本文中,我们提出了一种新的深度度量学习方法,该方法结合了三元组模型和嵌入空间的全局结构。我们依赖于一个智能挖掘过程,以低计算成本产生有效的训练样本。此外,我们提出了一种自适应控制器,可以自动调整智能挖掘超参数,加快训练过程的收敛速度。我们的经验表明,我们提出的方法可以比其他竞争挖掘方法更快,更准确地训练三元组ConvNets。此外,我们还证明了我们的方法在CUB-200-2011和Cars 196数据集上实现了新的最先进的嵌入结果。

英文摘要:

To solve deep metric learning problems and producing feature embeddings, current methodologies will commonly use a triplet model to minimise the relative distance between samples from the same class and maximise the relative distance between samples from different classes. Though successful, the training convergence of this triplet model can be compromised by the fact that the vast majority of the training samples will produce gradients with magnitudes that are close to zero. This issue has motivated the development of methods that explore the global structure of the embedding and other methods that explore hard negative/positive mining. The effectiveness of such mining methods is often associated with intractable computational requirements. In this paper, we propose a novel deep metric learning method that combines the triplet model and the global structure of the embedding space. We rely on a smart mining procedure that produces effective training samples for a low computational cost. In addition, we propose an adaptive controller that automatically adjusts the smart mining hyper-parameters and speeds up the convergence of the training process. We show empirically that our proposed method allows for fast and more accurate training of triplet ConvNets than other competing mining methods. Additionally, we show that our method achieves new state-of-the-art embedding results for CUB-200-2011 and Cars196 datasets.

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。

↑