Deep Feature Consistent Variational Autoencoder

Abstract

We present a novel method for constructing Variational Autoencoder (VAE). Instead of using pixel-by-pixel loss, we enforce deep feature consistency between the input and the output of a VAE, which ensures the VAE's output to preserve the spatial correlation characteristics of the input, thus leading the output to have a more natural visual appearance and better perceptual quality. Based on recent deep learning works such as style transfer, we employ a pre-trained deep convolutional neural network (CNN) and use its hidden features to define a feature perceptual loss for VAE training. Evaluated on the CelebA face dataset, we show that our model produces better results than other methods in the literature. We also show that our method can produce latent vectors that can capture the semantic information of face expressions and can be used to achieve state-of-the-art performance in facial attribute prediction.

WACV

References in corpus (9)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks
• Understanding Neural Networks Through Deep Visualization
• Perceptual Losses for Real-Time Style Transfer and Super-Resolution
• DRAW: A Recurrent Neural Network For Image Generation
• Autoencoding beyond pixels using a learned similarity metric
• Attribute2Image: Conditional Image Generation from Visual Attributes
• Combining Markov Random Fields and Convolutional Neural Networks for Image Synthesis
• Discriminative Regularization for Generative Models

Cited by in corpus (6)

• Variational Inference of Disentangled Latent Concepts from Unlabeled Observations
• Deep Feature Consistent Deep Image Transformations: Downscaling, Decolorization and HDR Tone Mapping
• A survey on Kornia: an Open Source Differentiable Computer Vision Library for PyTorch
• Explaining Predictions by Approximating the Local Decision Boundary
• A survey on Variational Autoencoders from a GreenAI perspective
• REX: Revisiting Budgeted Training with an Improved Schedule