Deep Learning with Vector Graphics
Deep Learning with Vector Graphics
About
This book
The author
Background knowledge
Vector Graphics
Scalable Vector Graphics (SVG) format
Deep Learning
Neural architectures
Deep generative models
Deep Learning with Vector Graphics
Arguing for E-2-E approach
Challenges of Vector Graphics
Strategies for dealing with SVG
Relevant work
Key academic publications
Alex Graves (2014) | Generating Sequences With Recurrent Neural Networks
David Ha (Dec 2015) | Handwriting Generation Demo in TensorFlow (blog post)
David Ha (Dec 2015) | Recurrent Net Dreams Up Fake Chinese Characters in Vector Format with TensorFlow (blog post)
Carter et al. (2016) | Four Experiments in Handwriting with a Neural Network (Distill publication)
Ha & Eck (2017) | A Neural Representation of Sketch Drawings (sketch-rnn)
Kimberli Zhong (2018) | Learning to Draw Vector Graphics: Applying Generative Modeling to Font Glyphs (Master thesis @ MIT)
Lopes et al. (2019) | A Learned Representation for Scalable Vector Graphics (svg-vae)
Carlier et al. (2020) | DeepSVG: A Hierarchical Generative Network for Vector Graphics Animation
Li et al. (2020) | Differentiable Vector Graphics Rasterization for Editing and Learning (diffvg)
Reddy et al. (2021) | Im2Vec: Synthesizing Vector Graphics without Vector Supervision
Wang and Lian (2021) | DeepVecFont: Synthesizing High-quality Vector Fonts via Dual-modality Learning
Vinker et al. (2022) | CLIPasso: Semantically-Aware Object Sketching
Ma et al. (2022) | LIVE: Towards Layer-wise Image Vectorization
Aoki and Aizawa (2022) | SVG vector font generation for Chinese Characters with Transformer
Jain et al. (2022) | VectorFusion: Text-to-SVG by Abstracting Pixel-Based Diffusion Models
Further academic publications
Campbell and Kautz (2014) | Learning a Manifold of Fonts
Das et al. (2020) | BézierSketch: A generative model for scalable vector sketches
Zhao et al. (2020) | ICONATE: Automatic Compound Icon Generation and Ideation
Das et al. (2021) | Cloud2Curve: Generation and Vectorization of Parametric Sketches
Yang et al. (2021) | SketchAA: Abstract Representation for Abstract Sketches
Jiang et al. (2021) | Recognizing Vector Graphics without Rasterization
Other relevant work
Sam Greydanus (2016) | Scribe: Generating Realistic Handwriting with TensorFlow
Google’s “Quick, Draw!” (2017)
Nicolas Boillot’s Vectoglyph (2019)
GAN XML Fixer (2019)
Datasets
Data representation
From strokes to SVG
Common SVG preprocessing
First steps
“De-minifying” path data
Harmonizing command positioning logic
Decomposition of basic SVG shapes to paths
Reducing the set of different path commands
Normalization
Path simplification
Finishing touches
Embedding
Open questions
Open questions
Acknowledgements
Acknowledgements
repository
open issue
Index
