CVAE
- Generative model with conditional information (labels or attributes)
- Training
input: sample X and condition Y
output: X
- Testing
input: condition Y and Gaussian noise vector z
output: reconstructed result
Examples
- Attribute-conditioned image generation
- Learning diverse image colorization
- Forecasting from static images
- Facial expression editting
Reference
- Yan X, Yang J, Sohn K, et al. Attribute2Image: Conditional Image Generation from Visual Attributes[J]. european conference on computer vision, 2015: 776-791.
- Deshpande A, Lu J, Yeh M C, et al. Learning Diverse Image Colorization[J]. 2016.
- Walker J, Doersch C, Gupta A, et al. An Uncertain Future: Forecasting from Static Images Using Variational Autoencoders[M]// Computer Vision – ECCV 2016. Springer International Publishing, 2016.
- Yeh R, Liu Z, Goldman D B, et al. Semantic Facial Expression Editing using Autoencoded Flow[J]. 2016.
- Doersch C. Tutorial on Variational Autoencoders[J]. 2016.
