Improving boundary quality in multi-region image stylization
Enhancing Multi-Region Stylization with Interior-Guided Boundary Repair
Computer Vision and Pattern Recognition
Summary
When transforming different parts of a picture into artistic styles independently, the edges where regions meet often look messy. The authors propose a simple method that fixes these edges by using information from inside each region to repair boundary pixels. Their approach blends colors near object edges carefully, so styles from nearby objects do not mix incorrectly. They show their method works better than previous ways to make these boundaries look smooth and natural.
What this means in practice
- •For digital artists: Ensure cleaner edges when styling individual regions in artwork created with neural style transfer tools.
- •For mobile app developers: Enhance photo editing apps by integrating improved boundary handling for multi-region artistic filters without retraining AI models.$Commercial implications: Allows selling apps with superior stylization quality by providing high-quality edge repair that existing AI style transfer tools lack.
Authors
Hong-Son Nguyen, Thi-Ngoc-Hanh Le
Abstract
Region-based neural style transfer enables fine-grained artistic control by allowing independent stylization of semantic image regions. However, compositing these regions often leads to boundary artifacts, degrading visual quality. We propose Interior-Guided Boundary Repair (IGBR), a lightweight and model-agnostic method that improves boundary handling in multi-region stylization. IGBR repairs boundary pixels using interior-guided propagation and applies inward, distance-based blending restricted to object-background boundaries, preventing inter-object style leakage. The method is derived from a region-wise constrained formulation with a closed-form solution and can be seamlessly integrated into existing stylization pipelines without retraining. To evaluate efficiency of our IGBR, we introduce quantitative metrics that measure boundary consistency, gradient artifacts, inter-object leakage, and interior preservation without requiring annotated stylized images. Our experiments and evaluations demonstrate that the proposed IGBR consistently produces plausible boundaries, outperforming prior blending techniques in boundary consistency, gradient stability, and interior preservation. The code is available at https://github.com/Son-SDT/IGBR.