Research map / Image generation / Editing and style transfer
Multimodal Transfer: A Hierarchical Deep Convolutional Neural Network for Fast Artistic Style Transfer
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Use hierarchical multimodal transfer instead of singular transfer trained at style size 256 | MT Net · High-resolution artistic style transfer | visual similarity to the artistic style | Helped |
| Use hierarchical multimodal transfer instead of singular transfer trained at style size 1024 | MT Net · High-resolution artistic style transfer | capture of texture distortion and fine brushwork | Helped |
| Use hierarchical transfer rather than a same-structure deep singular network | 1024 × 1024 image generation · 1024 × 1024 images | test time: 0.63s -> 0.54s | Helped |
| Use hierarchical transfer rather than a same-structure deep singular network | 1024 × 1024 image generation · 1024 × 1024 images | memory usage: ~6700 MB -> ~3100 MB | Helped |
| Train the singular transfer network with style size 256 rather than the multimodal hierarchy | Singular transfer · High-resolution artistic style transfer | texture scale match | Hurt |
| Train the singular transfer network with style size 1024 rather than the multimodal hierarchy | Singular transfer · High-resolution artistic style transfer | capture of texture distortion and fine brushwork | Hurt |
| Vary the hierarchical stylization-loss weights λk (ablation) | MT Net · Artistic style transfer | stylization result quality | No clear effect |
| Combine two styles, assigning coarse texture to one and fine brushwork to the other | MT Net · Two-style transfer | fusion of distinct style cues | Helped |
Closest papers by meaning
- A Flexible Convolutional Solver for Fast Style Transfers (CVPR 2019 · similarity 0.93)
- Deep Photo Style Transfer (CVPR 2017 · similarity 0.91)
- Artistic Style Transfer with Internal-external Learning and Contrastive Learning (NeurIPS 2021 · similarity 0.91)
- ArtBank: Artistic Style Transfer with Pre-trained Diffusion Model and Implicit Style Prompt Bank (AAAI 2024 · similarity 0.89)
- Style Injection in Diffusion: A Training-free Approach for Adapting Large-scale Diffusion Models for Style Transfer (CVPR 2024 · similarity 0.88)
- Puff-Net: Efficient Style Transfer with Pure Content and Style Feature Fusion Network (CVPR 2024 · similarity 0.88)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.