构建1.5万对数据集，提升手绘3D动画新视角还原质量
SPECSIA: Stylization Dataset for Novel-View Enhancement in Drawing-based 3D Animation
- 用14,980对图像训练去噪模块，修复新视角投影缺陷
- 新视角保真度与时间连贯性显著提升，单角色适配成本更低
- 适合做手绘3D动画生成的开发者和创作者使用
从单张2D手绘图生成动画极具挑战，需在保持角色外观的同时确保动作合理且时间连贯。现有方法多依赖逐样本2D优化来对齐动画渲染与输入图像，但易过拟合于观察视角，无法修正新视角中的投影伪影。为此，我们提出SPECSIA-15K数据集，包含1,498个3DBiCar角色的14,980对带伪影的投影/精修目标图像对。进一步设计DraViE（Drawing-based View Enhancement）轻量级插件模块，基于数据先验训练，可在保留风格与动作合理性的同时消除新视角伪影。实验表明，该方法在新视角保真度与时间连贯性上均有稳定提升，且单角色适配成本低于逐样本微调。
原文摘要 · Abstract (English)
Generating animation from a single 2D drawing is challenging because the output must preserve character appearance while remaining plausible and temporally coherent under motion. Existing drawing-based 3D animation pipelines often use sample-wise 2D refinement to align animated renderings with the input image, but such optimization tends to overfit to the observed view and fails to correct projection-induced artifacts in novel views. To address this limitation, we introduce SPECSIA-15K, a paired stylization dataset containing 14,980 artifact-corrupted projection/refinement-target pairs from 1,498 3DBiCar characters. We further present DraViE (Drawing-based View Enhancement), a lightweight plug-and-play module trained with data-level priors to remove novel-view artifacts while preserving style and motion plausibility. Experiments show consistent gains in novel-view fidelity and temporal coherence with lower per-character adaptation cost than sample-wise fine-tuning.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。