通过循环一致掩码预测学习跨视角物体对应
Learning Cross-View Object Correspondence via Cycle-Consistent Mask Prediction
提出基于条件二值分割的跨视角物体对应框架，利用循环一致性作为自监督信号，并支持测试时训练，可复用于视频中的物体跟踪与跨视角定位任务。
原文依据 We propose a simple yet effective framework based on conditional binary segmentation, where an object query mask is encoded into a latent representation to guide the localization of the corresponding object in a target video.
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论我们研究了在视频中建立跨不同视角的物体级视觉对应关系的任务，重点关注具有挑战性的自我中心到外中心和外中心到自我中心的场景。我们提出了一个简单而有效的基于条件二值分割的框架，其中物体查询掩码被编码为潜变量表示，以指导目标视频中对应物体的定位。为了促进鲁棒且视角不变的表示，我们引入了循环一致性训练目标：将目标视图中的预测掩码投影回源视图以重建原始查询掩码。这种双向约束提供了强大的自监督信号，无需真实标注，并能在推理时进行测试时训练（TTT）。在Ego-Exo4D和HANDAL-X基准上的实验证明了我们的优化目标和TTT策略的有效性，达到了最优性能。代码可在https://github.com/shannany0606/CCMP获取。
英文原文摘要 1047 字符
We study the task of establishing object-level visual correspondence across different viewpoints in videos, focusing on the challenging egocentric-to-exocentric and exocentric-to-egocentric scenarios. We propose a simple yet effective framework based on conditional binary segmentation, where an object query mask is encoded into a latent representation to guide the localization of the corresponding object in a target video. To encourage robust, view-invariant representations, we introduce a cycle-consistency training objective: the predicted mask in the target view is projected back to the source view to reconstruct the original query mask. This bidirectional constraint provides a strong self-supervisory signal without requiring ground-truth annotations and enables test-time training (TTT) at inference. Experiments on the Ego-Exo4D and HANDAL-X benchmarks demonstrate the effectiveness of our optimization objective and TTT strategy, achieving state-of-the-art performance. The code is available at https://github.com/shannany0606/CCMP.