2020/02/03 by Peng Xu, Xu, Peng, Zeyu Song +7 · 3 citations
Computer Science · #Advanced Image and Video Retrieval Techniques #Artificial intelligence #Computer Vision and Pattern Recognition (cs.CV) #Computer science #Deep learning #Exploit #FOS: Computer and information sciences #Feature (linguistics) #Feature learning #Human Pose and Action Recognition #Key (lock) #Machine learning #Multimodal Machine Learning Applications #Representation (politics) #Set (abstract data type) #Sketch #Sketch recognition #cs.CV
paper · pdf · doi:10.48550/arxiv.2002.00867
published in arXiv (Cornell University) (Cornell University)
arxiv created 2020/02/03 · openalex publication_date 2020/02/03 · arxiv updated 2020/02/04 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/05
In this paper, we tackle for the first time, the problem of self-supervised representation learning for free-hand sketches. This importantly addresses a common problem faced by the sketch community -- that annotated supervisory data are difficult to obtain. This problem is very challenging in that sketches are highly abstract and subject to different drawing styles, making existing solutions tailored for photos unsuitable. Key for the success of our self-supervised learning paradigm lies with our sketch-specific designs: (i) we propose a set of pretext tasks specifically designed for sketches that mimic different drawing styles, and (ii) we further exploit the use of a textual convolution network (TCN) in a dual-branch architecture for sketch feature learning, as means to accommodate the sequential stroke nature of sketches. We demonstrate the superiority of our sketch-specific designs through two sketch-related applications (retrieval and recognition) on a million-scale sketch dataset, and show that the proposed approach outperforms the state-of-the-art unsupervised representation learning methods, and significantly narrows the performance gap between with supervised representation learning.