O-MaMa：学习自我中心视角与外视角之间的对象掩码匹配
O-MaMa: Learning Object Mask Matching between Egocentric and Exocentric Views
AI总结：
O-MaMa将跨视角分割重构为掩码匹配任务，结合掩码上下文编码器、交叉注意力与对比损失，在Ego-Exo4D基准上以极低参数实现最先进性能。
摘要：
从多个视角理解世界对于协同工作的智能系统至关重要，其中在不同视角间分割共同对象仍是一个开放问题。我们提出了一种新方法，将跨图像分割重新定义为掩码匹配任务。我们的方法包括：（1）掩码上下文编码器，从FastSAM掩码候选中池化密集的DINOv2语义特征，以获得具有判别性的对象级表示；（2）自我中心↔外中心交叉注意力，融合多视角观测；（3）掩码匹配对比损失，在共享潜在空间中对齐跨视角特征；（4）硬负样本邻近挖掘策略，鼓励模型更好地区分邻近对象。O-MaMa在Ego-Exo4D对应关系基准上达到了最先进水平，在Ego2Exo和Exo2Ego的IoU上相对于官方挑战基线分别获得了+22%和+76%的相对提升，并且与使用1%训练参数的最先进方法相比，分别提升了+13%和+6%。
英文摘要：
Understanding the world from multiple perspectives is essential for intelligent systems operating together, where segmenting common objects across different views remains an open problem. We introduce a new approach that re-defines cross-image segmentation by treating it as a mask matching task. Our method consists of: (1) A Mask-Context Encoder that pools dense DINOv2 semantic features to obtain discriminative object-level representations from FastSAM mask candidates, (2) an Ego$\leftrightarrow$Exo Cross-Attention that fuses multi-perspective observations, (3) a Mask Matching contrastive loss that aligns cross-view features in a shared latent space, and (4) a Hard Negative Adjacent Mining strategy to encourage the model to better differentiate between nearby objects. O-MaMa achieves the state of the art in the Ego-Exo4D Correspondences benchmark, obtaining relative gains of +22% and +76% in the Ego2Exo and Exo2Ego IoU against the official challenge baselines, and a +13% and +6% compared with the SOTA with 1% of the training parameters.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
