发表机构
South China University of Technology（华南理工大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
LIFD通过锚定扩散与多视图一致性学习三维感知场景记忆，从单视图和循环记忆完成场景表示，提升机器人操作成功率，在LIBERO和MetaWorld上分别达91.6%和79.8%。
AI 中文摘要
在部分可观测条件下的机器人操作需要超出当前视野的空间信息。几何感知的RGB特征描述了可见结构，但随着机器人或场景的移动，先前观察到的区域可能消失。因此，维持有用的场景表示需要在推断缺失内容的同时保留观察历史，且不丢失与可见证据的联系。我们引入了LIFD（观察、想象、聚焦与执行），一个用于持久、三维感知场景记忆的框架。LIFD从多视图一致性中学习场景令牌表示，并从单一RGB视图和循环记忆中完成该表示。一个修正流模型生成令牌，而锚定引导的交叉注意力将完成过程条件化于当前几何特征。紧凑的槽位特征将该表示连接到操作策略。表示学习期间使用多视图和几何监督；部署时仅需一个RGB相机、本体感觉和任务指令。LIFD（分阶段）在LIBERO上达到91.6%的平均成功率，在MetaWorld上达到79.8%，相比联合训练在LIBERO上的平均成功率提高了3.1个百分点。在四个UR5e任务族（每个族十个演示）上，它实现了56.0%的平均成功率，而OpenVLA-7B为40.5%。
英文摘要
During manipulation, robot and scene motion can move previously observed regions outside the camera's field of view. Geometry-aware RGB features encode visible structure, while control under partial observability requires scene memory that integrates observation history and grounds inferred content in current evidence. We introduce \lifd{} (Look, Imagine, Focus, and Do), a framework for persistent, 3D-aware scene memory. LIFD learns scene tokens through multi-view agreement, then completes them from a single RGB view and recurrent memory using rectified flow. Anchor-Guided Cross-Attention anchors generation to current geometry-aware features, and compact slot features condition a visuomotor policy. Multi-view and geometric supervision are used during representation learning; deployment requires one RGB camera, proprioception, and a task instruction. LIFD (Staged) reaches 91.6\% average success on LIBERO and 79.8\% on MetaWorld, improving LIBERO average success by 11.1 percentage points over Joint training. After policy-head adaptation with ten demonstrations per family, LIFD achieves 56.0\% mean success across four UR5e task families, compared with 40.5\% for OpenVLA-7B.
Comments8 pages, 4 figures. Submitted to ICRA 2027
