基于属性下降的内容一致性车辆数据集仿真
Simulating Content Consistent Vehicle Datasets with Attribute Descent
摘要：
本文利用图形引擎模拟了大量带有自由标注的训练数据。在合成数据和真实数据之间，存在两级域间隙，即，内容层次和外观层次。虽然后者已被广泛研究，但我们专注于减少照明和视点等属性的内容差距。为了降低问题的复杂性，我们选择了一个更小，更可控的应用程序，车辆重新识别（重新ID）。我们介绍了一个大规模的合成数据集VehicleX。它在Unity中创建，包含1，362辆各种3D模型的车辆，具有完全可编辑的属性。我们提出了一种属性下降方法，让VehicleX近似真实世界数据集中的属性。具体来说，我们操纵VehicleX中的每个属性，旨在最大限度地减少VehicleX和真实数据之间在Fréchet初始距离（FID）方面的差异。该属性下降算法允许与现有外观DA方法正交的内容领域自适应（DA）。我们将优化的VehicleX数据与真实世界的车辆re-ID数据集混合，并观察到一致的改进。通过增强的数据集，我们报告了具有竞争力的准确性。我们在https://github.com/yorkeyao/VehicleX上提供数据集、引擎和代码。
英文摘要：
This paper uses a graphic engine to simulate a large amount of training data with free annotations. Between synthetic and real data, there is a two-level domain gap, i.e., content level and appearance level. While the latter has been widely studied, we focus on reducing the content gap in attributes like illumination and viewpoint. To reduce the problem complexity, we choose a smaller and more controllable application, vehicle re-identification (re-ID). We introduce a large-scale synthetic dataset VehicleX. Created in Unity, it contains 1,362 vehicles of various 3D models with fully editable attributes. We propose an attribute descent approach to let VehicleX approximate the attributes in real-world datasets. Specifically, we manipulate each attribute in VehicleX, aiming to minimize the discrepancy between VehicleX and real data in terms of the Fréchet Inception Distance (FID). This attribute descent algorithm allows content domain adaptation (DA) orthogonal to existing appearance DA methods. We mix the optimized VehicleX data with real-world vehicle re-ID datasets, and observe consistent improvement. With the augmented datasets, we report competitive accuracy. We make the dataset, engine and our codes available at https://github.com/yorkeyao/VehicleX.
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。
