SPECSIA：用于基于绘图的3D动画中新颖视角增强的风格化数据集
SPECSIA: Stylization Dataset for Novel-View Enhancement in Drawing-based 3D Animation
- School of Electrical Engineering, KAIST, Daejeon, Republic of Korea（韩国成均馆大学电子工程学院）
- Department of AI, Chung-Ang University, Seoul, Republic of Korea（韩国 Chung-Ang 大学人工智能系）
机构由 AI 辅助整理，请以论文原文为准。
AI总结：
提出SPECSIA-15K数据集和DraViE模块，通过数据先验去除新颖视角伪影，提升动画保真度和时间一致性，降低每角色适应成本。
AI中文摘要：
从单个2D绘图生成动画具有挑战性，因为输出必须在运动下保持角色外观的同时保持合理性和时间一致性。现有的基于绘图的3D动画流程通常使用样本级2D细化来对齐动画渲染与输入图像，但这种优化往往过度拟合观察视角，无法纠正新颖视角下由投影引起的伪影。为解决这一限制，我们引入了SPECSIA-15K，一个配对的风格化数据集，包含来自1,498个3DBiCar角色的14,980个伪影损坏的投影/细化目标对。我们进一步提出了DraViE（基于绘图的视角增强），一个轻量级的即插即用模块，通过数据级先验训练，在保持风格和运动合理性的同时去除新颖视角伪影。实验表明，与样本级微调相比，该方法在新颖视角保真度和时间一致性上持续提升，且每角色适应成本更低。
英文摘要：
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.
