MHE-Former：基于熵最大化的多假设Transformer用于3D网格恢复
原标题：MHE-Former: Multi-Hypothesis Transformers via Entropy Maximization for 3D Mesh Recovery
AI 摘要
该论文提出 MHE-Former，一种基于熵最大化的多假设 Transformer 框架，用于单目 3D 手部和身体网格恢复。它采用探索-利用范式：探索阶段通过概率建模和熵最大化生成多样且合理的假设，利用阶段借助 VLM 的视觉理解与推理能力进行上下文感知的假设选择。实验表明该框架在多个数据集上取得准确性和多样性的最优表现，用户偏好研究也验证了假设选择流程的实用性。
正文节选
MHE-Former: Multi-Hypothesis Transformers via Entropy Maximization for 3D Mesh Recovery Abstract Monocular 3D hand and body mesh recovery often suffers from severe occlusion and ambiguity. Traditional deterministic methods typically regress a single optimal solution, leading to overconfident predictions. In this paper, we introduce an exploration–exploitation paradigm for ambiguous mesh recovery with multi-hypothesis learning and selection. Specifically, during exploration, based on our probabil