State-of-the-Art survey of deep learning based sketch retrieval

Sketch based image retrieval (SBIR) is an important research issue in computer vision, which is a flexible and convenient way to retrieve target images. How to minimize the difference between the sketch domain and the image domain is the key to the SBIR method. Deep learning methods have used in image processing research widely, which break through the limitations of traditional methods, which extract high-dimensional features from a large amount of data and have been proved to effectively solve the cross-domain modeling issues. The research focuses on the deep learning-based sketch retrieval method and reviews the related research work, in the aspects of deep feature extraction model, coarse-grained and fine-grained retrieval based on deep learning, category generalization. Finally, the challenges and future research directions of sketch retrieval are summarized and prospected.