WebtoonMe: A Data-Centric Approach for Full-Body Portrait Stylization
This work addresses the challenge of creating production-level cartoon-style full-body portraits, which is incremental as it builds on a two-stage method with a novel dataset preparation approach.
The paper tackles the problem of full-body portrait stylization by proposing a data-centric solution that improves output plausibility and quality robustness for non-face regions, achieving high-quality stylization without extra losses or architectural changes.
Full-body portrait stylization, which aims to translate portrait photography into a cartoon style, has drawn attention recently. However, most methods have focused only on converting face regions, restraining the feasibility of use in real-world applications. A recently proposed two-stage method expands the rendering area to full bodies, but the outputs are less plausible and fail to achieve quality robustness of non-face regions. Furthermore, they cannot reflect diverse skin tones. In this study, we propose a data-centric solution to build a production-level full-body portrait stylization system. Based on the two-stage scheme, we construct a novel and advanced dataset preparation paradigm that can effectively resolve the aforementioned problems. Experiments reveal that with our pipeline, high-quality portrait stylization can be achieved without additional losses or architectural changes.