面向Logo识别的合成数据生成方法研究

Research on synthesis data generation method for logo recognition

针对深度学习框架下Logo识别任务中可训练样本稀疏的问题,提出了一种基于上下文的Logo数据合成算法,该算法综合利用了Logo对象内部、Logo周围邻域、Logo与其他对象之间以及Logo所处场景等多种类型的上下文信息指导Logo图像的合成。在Flickr Logos-32数据集上的实验结果显示,所提算法能够在不依赖额外手工标注的前提下,提升Logo识别算法的性能(mA P提升8.5%),验证了该合成算法的有效性。

Aiming at the problem of training sample sparse in Logo recognition task under the deep learning framework, a Logo data synthesis algorithm based on contexts was proposed. The algorithm comprehensively utilizes various types of context information to guide the synthesis of Logo images, such as the interior of Logo object, the neighborhood of Logo object, the link between Logo object and other objects and the scene where Logo object lives in. The experimental results on the Flickr Logos-32 dataset show that the proposed algorithm can improve the performance of the Logo identification algorithm（m AP increase by 8.5%） without relying on additional manual annotation, verifying the effectiveness of the synthesis algorithm.