Learning Inclusion Matching for Animation Paint Bucket Colorization

Colorizing line art is a pivotal task in the production of hand-drawn cel animation. Digital painters use a paint bucket tool to manually color each segment enclosed by lines, based on RGB values predetermined by color designers. This frame-by-frame process, named Paint Bucket Colorization, is both arduous and time-intensive.

Current automated colorization methods mainly focus on segment matching. This technique migrates colors from a reference to the target frame by aligning features within line-enclosed segments across frames. However, issues like occlusion and wrinkles in animations often disrupt these direct correspondences, leading to mismatches. In this work, we introduce a new learning-based inclusion matching pipeline, which directs the network to comprehend the inclusion relationships between segments rather than relying solely on direct visual correspondences. Our method features a two-stage estimation module that integrates a coarse color warping module with an inclusion matching module, enabling more nuanced and accurate colorization.

To facilitate the training of this network, we have developed a unique dataset, referred to as PaintBucket-Character. This dataset includes rendered line arts alongside their colorized counterparts, featuring various 3D characters. Extensive experiments demonstrate the effectiveness and superiority of our method over existing techniques.