Style-CCL: Content-Preserving Style Transfer via Curriculum Continual Learning
This work addresses the challenge of content-preserving style transfer for Diffusion Transformers, where existing methods struggle with entangled features and style dominance.
Style-CCL introduces a curriculum continual learning framework for Diffusion Transformers that decouples content and style, achieving state-of-the-art performance in style similarity, content consistency, and aesthetic quality.
Content-Preserving Style transfer, given content and style references, remains challenging for Diffusion Transformers (DiTs) due to entangled content and style features. With a reverse triplet synthesis pipeline to build a million-scale training set and a dual-branch Style-Content DiT (SC-DiT) that decouples style and content via separate ROPE embeddings and causal masking, we observe that such a one-stage training paradigm on mixed style categories causes semantic styles to dominate, hindering texture style learning, and harming content preservation. To address these issues, we propose Style-CCL, a Multi-Stage Curriculum Continual Learning framework that trains SC-DiT from semantic (easy) to texture (hard) styles, and from clean to synthetic data, with Random Memory Rehearsal across stages to avoid catastrophic forgetting. Extensive experiments demonstrate that our Style-CCL achieves state-of-the-art performance in three core metrics: style similarity, content consistency, and aesthetic quality.