Adversarial Training For Sketch Retrieval
This addresses retrieval of sketch-like symbols in heritage documents, but it is incremental as it adapts GANs to a new task.
The paper tackled the problem of applying GAN-learned representations to retrieval, specifically for unlabelled Merchant Marks in heritage documents, and found that their novel sketch-GAN architecture achieved increased stability to rotation, scale, and translation compared to standard GANs.
Generative Adversarial Networks (GAN) are able to learn excellent representations for unlabelled data which can be applied to image generation and scene classification. Representations learned by GANs have not yet been applied to retrieval. In this paper, we show that the representations learned by GANs can indeed be used for retrieval. We consider heritage documents that contain unlabelled Merchant Marks, sketch-like symbols that are similar to hieroglyphs. We introduce a novel GAN architecture with design features that make it suitable for sketch retrieval. The performance of this sketch-GAN is compared to a modified version of the original GAN architecture with respect to simple invariance properties. Experiments suggest that sketch-GANs learn representations that are suitable for retrieval and which also have increased stability to rotation, scale and translation compared to the standard GAN architecture.