Abstract
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.
Keywords
Subject
Publication details
- Journal
- Not available
- Open access
- Green open access
Cite this article
APA 7
Li, W., Chen, Y., Li, W., & Wu, Q. (2026). LIFD: Anchored Diffusion for 3D-Aware Scene Memory in Robotic Manipulation. https://omanscience.com/en/articles/lifd-anchored-diffusion-for-3d-aware-scene-memory-in-robotic-manipulation
MLA 9
Li, Wenbo, et al. "LIFD: Anchored Diffusion for 3D-Aware Scene Memory in Robotic Manipulation." https://omanscience.com/en/articles/lifd-anchored-diffusion-for-3d-aware-scene-memory-in-robotic-manipulation.
Chicago (author–date)
Li, Wenbo, Yiteng Chen, Wenhao Li, and Qingyao Wu. 2026. "LIFD: Anchored Diffusion for 3D-Aware Scene Memory in Robotic Manipulation." https://omanscience.com/en/articles/lifd-anchored-diffusion-for-3d-aware-scene-memory-in-robotic-manipulation.
Harvard
Li, W., Chen, Y., Li, W. and Wu, Q. (2026) 'LIFD: Anchored Diffusion for 3D-Aware Scene Memory in Robotic Manipulation', Available at: https://omanscience.com/en/articles/lifd-anchored-diffusion-for-3d-aware-scene-memory-in-robotic-manipulation.
Vancouver
Li W, Chen Y, Li W, Wu Q. LIFD: Anchored Diffusion for 3D-Aware Scene Memory in Robotic Manipulation. https://omanscience.com/en/articles/lifd-anchored-diffusion-for-3d-aware-scene-memory-in-robotic-manipulation
IEEE
W. Li, Y. Chen, W. Li, and Q. Wu, "LIFD: Anchored Diffusion for 3D-Aware Scene Memory in Robotic Manipulation," https://omanscience.com/en/articles/lifd-anchored-diffusion-for-3d-aware-scene-memory-in-robotic-manipulation.