“Confidence Estimation of Few-Shot Patch-Based Learning for Anime-Style Colorization” by Ji, Maejima, Sechayk, Koyama and Igarashi
Conference:
Type(s):
Title:
- Confidence Estimation of Few-Shot Patch-Based Learning for Anime-Style Colorization
Session/Category Title:
- Images, Video & Computer Vision
Presenter(s)/Author(s):
Abstract:
This study proposes a region-wise confidence estimation method for anime-style line drawing colorization. By comparing local patches in the colorized image with training images using normalized cross-correlation, the method highlights uncertain regions. It improves usability by aiding artists in identifying colorization errors efficiently and reliably.
References:
[1] Yu Cao, Xiangqiao Meng, PY Mok, Xueting Liu, Tong-Yee Lee, and Ping Li. 2023a. Animediffusion: Anime face line drawing colorization via diffusion models. arXiv preprint arXiv:https://arXiv.org/abs/2303.11137 (2023).
[2] Yu Cao, Hao Tian, and PY Mok. 2023b. Attention-aware anime line drawing colorization. In 2023 IEEE International Conference on Multimedia and Expo (ICME). IEEE, 1637–1642.
[3] Zhiheng Liu, Ka Leong Cheng, Xi Chen, Jie Xiao, Hao Ouyang, Kai Zhu, Yu Liu, Yujun Shen, Qifeng Chen, and Ping Luo. 2025. MangaNinja: Line Art Colorization with Precise Reference Following. arXiv preprint arXiv:https://arXiv.org/abs/2501.08332 (2025).
[4] Akinobu Maejima, Seitaro Shinagawa, Hiroyuki Kubo, Takuya Funatomi, Tatsuo Yotsukura, Satoshi Nakamura, and Yasuhiro Mukaigawa. 2024. Continual few-shot patch-based learning for anime-style colorization. Computational Visual Media 10, 4 (2024), 705–723.
