Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

VectorArk: Learning Practical Image Vectorization with Rounded Polygon Representation

About

Recent vision-language model (VLM)-based approaches have achieved impressive results on image vectorization tasks. However, they are typically evaluated on synthetic benchmarks, where clean SVGs are rasterized at high resolution and then re-vectorized. As a result, these methods generalize poorly to real-world scenarios, such as images with unknown rasterization methods or those generated by text-to-image models. We introduce VectorArk, a new VLM-based model designed for robust and practical image vectorization. VectorArk employs a novel rounded polygon representation that simplifies the learning process while naturally producing smooth, visually appealing primitives. We also propose a degradation model that enhances robustness across diverse and imperfect inputs. Our experiments show that, in contrast to previous methods, VectorArk achieves superior geometric completeness and artifact suppression across multiple datasets, with comprehensive ablations validating the contribution of each component.

Tarun Gehlaut, Difan Liu, Charu Bansal, Krutik Malani, Souymodip Chakraborty, Ankit Phogat, Matthew Fisher, Vineet Batra• 2026

Related benchmarks

TaskDatasetResultRank
Image VectorizationSVG-Stack-Simple (Easy)
LPIPS0.0274
10
Image VectorizationSVG-Fonts Simple (Easy)
LPIPS0.0217
10
Image VectorizationSVG-Icons-Simple (Easy)
LPIPS0.0565
10
Image VectorizationSVG-Emoji-Simple (Easy)
LPIPS0.0282
10
Image VectorizationMMSVG-Icon Easy (test)
LPIPS0.0122
5
Image VectorizationMMSVG-Icon Med (test)
LPIPS0.0179
5
Image VectorizationMMSVG-Icon Hard (test)
LPIPS0.0324
5
Image VectorizationSVGX core 250k Easy (test)
LPIPS0.0168
5
Image VectorizationSVGX core 250k Med (test)
LPIPS0.0332
5
Image VectorizationSVGX_core_250k Hard (test)
LPIPS0.0677
5
Showing 10 of 35 rows

Other info

Follow for update