探索自监督/半监督/全监督预训练模型上的数据增强
Exploring Data Augmentations on Self-/Semi-/Fully- Supervised Pre-trained Models
AI总结：
本文实证研究 Random Erasing、CutOut、CutMix 和 MixUp 四种数据增强对自监督/半监督/全监督预训练模型的影响，揭示遮盖区域降低不变性但提升多样性，而 MixUp 在几乎不损失不变性的情况下显著增强多样性。
摘要：
数据增强已成为视觉预训练模型的标准组成部分，用于捕获增强视图之间的不变性。在实践中，以零值/均值或来自其他样本的图像块遮盖样本区域的技术，通常被用于具有自监督/半监督/全监督对比损失的预训练模型中。然而，这些增强技术有效性背后的机制仍未得到充分探索。为研究这些问题，我们进行了一项实证研究，以量化数据增强如何影响性能。具体而言，我们将称为 Random Erasing、CutOut、CutMix 和 MixUp 的 4 种数据增强应用于一系列自监督/半监督/全监督预训练模型。我们报告了它们在图像分类、目标检测、实例分割和语义分割等视觉任务上的性能。然后，我们显式评估了特征嵌入的不变性和多样性。我们观察到：1）遮盖图像区域会降低所学特征嵌入的不变性，同时提供更可观的多样性。2）人工标注不会改变所学特征嵌入的不变性或多样性。3）MixUp 方法显著提高了多样性，而在不变性方面仅有轻微下降。
英文摘要：
Data augmentation has become a standard component of vision pre-trained models to capture the invariance between augmented views. In practice, augmentation techniques that mask regions of a sample with zero/mean values or patches from other samples are commonly employed in pre-trained models with self-/semi-/fully-supervised contrastive losses. However, the underlying mechanism behind the effectiveness of these augmentation techniques remains poorly explored. To investigate the problems, we conduct an empirical study to quantify how data augmentation affects performance. Concretely, we apply 4 types of data augmentations termed with Random Erasing, CutOut, CutMix and MixUp to a series of self-/semi-/fully- supervised pre-trained models. We report their performance on vision tasks such as image classification, object detection, instance segmentation, and semantic segmentation. We then explicitly evaluate the invariance and diversity of the feature embedding. We observe that: 1) Masking regions of the images decreases the invariance of the learned feature embedding while providing a more considerable diversity. 2) Manual annotations do not change the invariance or diversity of the learned feature embedding. 3) The MixUp approach improves the diversity significantly, with only a marginal decrease in terms of the invariance.
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。
