Research
My research interests are loosely organized into two thrusts:
Building 3D Representations with Vision
How can we build representations of the world that are informative but lightweight enough for embodied agents?
Decision Making for Embodied Agents
How can we use novel 3D representations to effectively plan under real-world conditions?
-
-
Zero-Shot Multi-Object Shape Completion. S Iwase, K Liu, V Guizilini, A Gaidon, K Kitani, R Ambrus, S Zakharov. To appear at ECCV 2024. [pdf]
-
DiffusionNOCS: Managing Symmetry and Uncertainty in Sim2Real Multi-Modal Category-level Pose Estimation. T Ikeda, S Zakharov, T Ko, MZ Irshad, R Lee, K Liu, R Ambrus, K Nishiwaki. To appear at IROS 2024. [pdf]
-
-
-
-
-
-
-
Online High-Level Model Estimation for Efficient Hierarchical Robot Navigation. M Stadler, K Liu, N Roy. IROS 2021. [pdf]
-
Hierarchical Object Map Estimation for Efficient and Robust Navigation. K Ok, K Liu, N Roy. ICRA 2021. [pdf]
-
-
-
-