QuantArt: Quantizing Image Style Transfer Towards High Visual Fidelity
Authors:
- 1 June 2023
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 5947-5956
Abstract
The mechanism of existing style transfer algorithms is by minimizing a hybrid loss function to push the generated image toward high similarities in both content and style. However, this type of approach cannot guarantee visual fidelity, i.e., the generated artworks should be indistinguishable from real ones. In this paper, we devise a new style transfer framework called QunntArt for high visual-fidelity stylization. QuantArt pushes the latent representation of the generated artwork toward the centroids of the real artwork distribution with vector quantization. By fusing the quantized and continuous latent representations, QuantArt allows flexible control over the generated artworks in terms of content preservation, style similarity, and visual fidelity. Experiments on various style transfer settings show that our QuantArt framework achieves significantly higher visual fidelity compared with the existing style transfer methods.Keywords
All Related Versions
Funding Information
- NIH (5U54CA225088-03)
This publication has 33 references indexed in Scilit:
- The Unreasonable Effectiveness of Deep Features as a Perceptual Metric2018
- Arbitrary Style Transfer in Real-Time with Adaptive Instance Normalization2017
- StyleBank: An Explicit Representation for Neural Image Style Transfer2017
- Improved Texture Networks: Maximizing Quality and Diversity in Feed-Forward Stylization and Texture Synthesis2017
- Deep Photo Style Transfer2017
- Diversified Texture Synthesis with Feed-Forward Networks2017
- Deep Residual Learning for Image Recognition2016
- Split and Match: Example-Based Adaptive Patch Sampling for Unsupervised Style Transfer2016
- ImageNet: A large-scale hierarchical image database2009
- Painterly rendering with curved brush strokes of multiple sizes1998