Y
Yifan Liu
Researcher at University of Adelaide
Publications - 22
Citations - 759
Yifan Liu is an academic researcher from University of Adelaide. The author has contributed to research in topics: Segmentation & Market data. The author has an hindex of 9, co-authored 21 publications receiving 469 citations. Previous affiliations of Yifan Liu include Beihang University.
Papers
More filters
Book ChapterDOI
Emotion Classification with Data Augmentation Using Generative Adversarial Networks
TL;DR: This paper designs a framework using a CNN model as the classifier and a cycle-consistent adversarial networks (CycleGAN) as the generator, and employs the least-squared loss as adversarial loss to avoid gradient vanishing problem.
Journal ArticleDOI
Structured Knowledge Distillation for Dense Prediction
TL;DR: Two structured distillation schemes are studied:pair-wise distillation that distills the pair-wise similarities by building a static graph; and holisticdistillation that uses adversarial training to distill holistic knowledge.
Journal ArticleDOI
Auto-painter: Cartoon image generation from sketch by using conditional Wasserstein generative adversarial networks
TL;DR: A model called auto-painter is proposed which can automatically generate compatible colors given a sketch and is not only capable of painting hand-draw sketch with compatible colors, but also allowing users to indicate preferred colors.
Posted Content
Data Augmentation in Emotion Classification Using Generative Adversarial Networks.
TL;DR: A data augmentation method using generative adversarial networks (GAN) that can complement and complete the data manifold and find better margins between neighboring classes and several evaluation methods to validate GAN's performance are proposed.
Posted Content
Auto-painter: Cartoon Image Generation from Sketch by Using Conditional Generative Adversarial Networks.
TL;DR: This paper proposes the auto-painter model which can automatically generate compatible colors for a sketch which is not only capable of painting hand-draw sketch with proper colors, but also allowing users to indicate preferred colors.