# Optimize and Reduce: A Top-Down Approach for Image Vectorization

arXiv: https://arxiv.org/abs/2312.11334 · HTML: https://mlsanity.com/paper/2312.11334/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 23 of its 53 references that are in the map.
- First posted on arXiv 18 Dec 2023.

## Builds on (most cited first, 23 of 23)

- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [Very Deep Convolutional Networks for Large-Scale Image Recognition](https://mlsanity.com/paper/1409.1556/) (2014)
- [The Unreasonable Effectiveness of Deep Features as a Perceptual Metric](https://mlsanity.com/paper/1801.03924/) (2018)
- [DreamFusion: Text-to-3D using 2D Diffusion](https://mlsanity.com/paper/2209.14988/) (2022)
- [CLIPDraw: Exploring Text-to-Drawing Synthesis through Language-Image Encoders](https://mlsanity.com/paper/2106.14843/) (2021)
- [DeepSVG: A Hierarchical Generative Network for Vector Graphics Animation](https://mlsanity.com/paper/2007.11301/) (2020)
- [VectorFusion: Text-to-SVG by Abstracting Pixel-Based Diffusion Models](https://mlsanity.com/paper/2211.11319/) (2022)
- [A Learned Representation for Scalable Vector Graphics](https://mlsanity.com/paper/1904.02632/) (2019)
- [Learning to generate line drawings that convey geometry and semantics](https://mlsanity.com/paper/2203.12691/) (2022)
- [Photo-Sketching: Inferring Contour Drawings from Images](https://mlsanity.com/paper/1901.00542/) (2019)
- [CLIPascene: Scene Sketching with Different Types and Levels of Abstraction](https://mlsanity.com/paper/2211.17256/) (2022)
- [Paint Transformer: Feed Forward Neural Painting with Stroke Prediction](https://mlsanity.com/paper/2108.03798/) (2021)
- [Stylized Neural Painting](https://mlsanity.com/paper/2011.08114/) (2020)
- [Modern Evolution Strategies for Creativity: Fitting Concrete Images and Abstract Concepts](https://mlsanity.com/paper/2109.08857/) (2021)
- [Learning Deep Sketch Abstraction](https://mlsanity.com/paper/1804.04804/) (2018)
- [CLIPVG: Text-Guided Image Manipulation Using Differentiable Vector Graphics](https://mlsanity.com/paper/2212.02122/) (2022)
- [De-rendering Stylized Texts](https://mlsanity.com/paper/2110.01890/) (2021)
- [ClipGen: A Deep Generative Model for Clipart Vectorization and Synthesis](https://mlsanity.com/paper/2106.04912/) (2021)
- [Differentiable Drawing and Sketching](https://mlsanity.com/paper/2103.16194/) (2021)
- [Im2Pencil: Controllable Pencil Illustration from Photographs](https://mlsanity.com/paper/1903.08682/) (2019)
- [Vectorization of Line Drawings via PolyVector Fields](https://mlsanity.com/paper/1801.01922/) (2018)
- [Inverse Diffusion Curves using Shape Optimization](https://mlsanity.com/paper/1610.02769/) (2016)

## Abstract (arXiv)

Vector image representation is a popular choice when editability and flexibility in resolution are desired. However, most images are only available in raster form, making raster-to-vector image conversion (vectorization) an important task. Classical methods for vectorization are either domain-specific or yield an abundance of shapes which limits editability and interpretability. Learning-based methods, that use differentiable rendering, have revolutionized vectorization, at the cost of poor generalization to out-of-training distribution domains, and optimization-based counterparts are either slow or produce non-editable and redundant shapes. In this work, we propose Optimize & Reduce (O&R), a top-down approach to vectorization that is both fast and domain-agnostic. O&R aims to attain a compact representation of input images by iteratively optimizing B'ezier curve parameters and significantly reducing the number of shapes, using a devised importance measure. We contribute a benchmark of five datasets comprising images from a broad spectrum of image complexities - from emojis to natural-like images. Through extensive experiments on hundreds of images, we demonstrate that our method is domain agnostic and outperforms existing works in both reconstruction and perceptual quality for a fixed number of shapes. Moreover, we show that our algorithm is × 10 faster than the state-of-the-art optimization-based method.