Contrastive View Design Strategies to Enhance Robustness to Domain Shifts in Downstream Object Detection
Contrastive View Design Strategies to Enhance Robustness to Domain Shifts in Downstream Object Detection
复制标题
DOI：
10.48550/arxiv.2212.04613
复制
发表时间：
2022-12
期刊：
影响因子：
--
通讯作者：
Kyle Buettner;Adriana Kovashka
中科院分区：
文献类型：
--
作者：
Kyle Buettner;Adriana Kovashka
对比学习已经成为一种有竞争力的对象检测预训练方法。尽管取得了这一进展，但在面对域偏移时，对对比预训练检测器的鲁棒性的研究很少。为了解决这一差距，我们进行了对比学习和域外对象检测的实证研究，研究对比视图设计如何影响鲁棒性。特别是，我们进行了以检测为重点的借口任务实例定位（InsLoc）的案例研究，并提出了策略，以增加视图，并提高鲁棒性的外观转移和上下文转移的情况下。在这些策略中，我们提出了对裁剪的更改，例如更改使用的百分比，添加IoU约束，以及集成基于显着性的对象先验。我们还探讨了增加捷径减少增强，如泊松混合，纹理平坦化和弹性变形。我们对这些策略进行了抽象，天气和上下文域转换的基准测试，并在单对象和多对象图像数据集的预训练中说明了将它们联合收割机结合起来的鲁棒方法。总的来说，我们的结果和见解显示了如何通过对比学习中的视图选择来确保鲁棒性。
Contrastive learning has emerged as a competitive pretraining method for object detection. Despite this progress, there has been minimal investigation into the robustness of contrastively pretrained detectors when faced with domain shifts. To address this gap, we conduct an empirical study of contrastive learning and out-of-domain object detection, studying how contrastive view design affects robustness. In particular, we perform a case study of the detection-focused pretext task Instance Localization (InsLoc) and propose strategies to augment views and enhance robustness in appearance-shifted and context-shifted scenarios. Amongst these strategies, we propose changes to cropping such as altering the percentage used, adding IoU constraints, and integrating saliency based object priors. We also explore the addition of shortcut-reducing augmentations such as Poisson blending, texture flattening, and elastic deformation. We benchmark these strategies on abstract, weather, and context domain shifts and illustrate robust ways to combine them, in both pretraining on single-object and multi-object image datasets. Overall, our results and insights show how to ensure robustness through the choice of views in contrastive learning.