Edge Augmentation for Large-Scale Sketch Recognition without Sketches
This addresses the problem of tedious sketch data collection for researchers in computer vision, offering a scalable method but with incremental improvements in domain adaptation.
The paper tackles the challenge of scaling sketch recognition to many categories by overcoming the lack of training sketch data through a novel augmentation technique that converts natural images into randomized edge representations, enabling training on 874 categories and demonstrating performance on a subset of 393 categories.
This work addresses scaling up the sketch classification task into a large number of categories. Collecting sketches for training is a slow and tedious process that has so far precluded any attempts to large-scale sketch recognition. We overcome the lack of training sketch data by exploiting labeled collections of natural images that are easier to obtain. To bridge the domain gap we present a novel augmentation technique that is tailored to the task of learning sketch recognition from a training set of natural images. Randomization is introduced in the parameters of edge detection and edge selection. Natural images are translated to a pseudo-novel domain called "randomized Binary Thin Edges" (rBTE), which is used as a training domain instead of natural images. The ability to scale up is demonstrated by training CNN-based sketch recognition of more than 2.5 times larger number of categories than used previously. For this purpose, a dataset of natural images from 874 categories is constructed by combining a number of popular computer vision datasets. The categories are selected to be suitable for sketch recognition. To estimate the performance, a subset of 393 categories with sketches is also collected.