研究人员开发了一个新的框架，可以从单张图像中重建高保真3D头部模型，保留面部身份，并通过粗到精的优化流程实现行业级拓扑。该方法结合了几何感知约束和辅助正则化来纠正精细伪影，用户研究表明专业艺术家发现结果几乎可用，并优于其他方法。另外，一个新的基准测试评估了各种视觉基础模型在检测AI生成图像方面的能力，发现最佳模型在准确率上比CLIP高出12%以上。这项工作还引入了一个可调注意力池化（TAP）机制，以更好地利用VFM特征，在AI生成图像检测领域建立了新的最先进水平。 AI
影响 3D头部建模的进步可以加速数字人类的创建，而改进的AI生成图像检测对于打击虚假信息和确保真实性至关重要。
排序理由 提出了两篇不同的研究论文，一篇关于3D头部建模，另一篇关于AI生成图像检测。
AI 生成摘要 · Google Gemini · 来自 4 个来源。 我们如何撰写摘要 →
完整方法见我们的编辑标准。
arXiv:2605.04524v1 Announce Type: new Abstract: We present a single-image head mesh reconstruction framework that addresses the longstanding challenge of simultaneously preserving facial identity and producing industry-grade topology. Our framework adopts a coarse-to-fine optimiz…
We present a single-image head mesh reconstruction framework that addresses the longstanding challenge of simultaneously preserving facial identity and producing industry-grade topology. Our framework adopts a coarse-to-fine optimization pipeline that refines a rigged template ac…
arXiv:2604.26772v1 Announce Type: new Abstract: Recent methods demonstrate that large-scale pretrained models, such as CLIP vision transformers, effectively detect AI-generated images (AIGIs) from unseen generative models when used as feature extractors. Many state-of-the-art met…
Recent methods demonstrate that large-scale pretrained models, such as CLIP vision transformers, effectively detect AI-generated images (AIGIs) from unseen generative models when used as feature extractors. Many state-of-the-art methods for AI-generated image detection build upon…