NoPA replaces Gaussian object approximations with non-parametric distributions and MMD-based merging to improve accuracy in real-time 3D scene graph generation.
In: 2020 IEEE-RAS 20th International Conference on Humanoid Robots (Humanoids)
6 Pith papers cite this work. Polarity classification is still indexing.
6
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
verdicts
UNVERDICTED 6roles
background 1polarities
background 1representative citing papers
VRA grounds discrete-time joint acceleration commands in voltage-constrained actuator physics to eliminate unrealizable accelerations and reduce oscillations in electric motor systems.
Play2Perfect uses task-agnostic RL play pretraining on diverse objects to build reusable manipulation priors, then fine-tunes for assembly, yielding 33x sample efficiency gains and 60% success on 0.5mm-clearance insertions in sim-to-real transfer.
Any2Any transfers humanoid whole-body tracking models across embodiments via kinematic alignment followed by targeted PEFT, matching full-training performance with 1% of the data and compute on tested platforms.
An empirical study of JEPA world models identifies architecture, training objective, and planning choices that yield a model outperforming DINO-WM and V-JEPA-2-AC on navigation and manipulation tasks.
MOBIUS is a multi-modal bipedal robot with hybrid reinforcement learning and force control plus an MIQCP planner that enables walking, crawling, climbing, and rolling on varied terrains.
citing papers explorer
-
NoPA: Non-Parametric Online 3D Scene Graph Generation
NoPA replaces Gaussian object approximations with non-parametric distributions and MMD-based merging to improve accuracy in real-time 3D scene graph generation.
-
VRA: Grounding Discrete-Time Joint Acceleration in Voltage-Constrained Actuation
VRA grounds discrete-time joint acceleration commands in voltage-constrained actuator physics to eliminate unrealizable accelerations and reduce oscillations in electric motor systems.
-
Play2Perfect: What Matters in Dexterous Play Pretraining for Precise Assembly?
Play2Perfect uses task-agnostic RL play pretraining on diverse objects to build reusable manipulation priors, then fine-tunes for assembly, yielding 33x sample efficiency gains and 60% success on 0.5mm-clearance insertions in sim-to-real transfer.
-
Any2Any: Efficient Cross-Embodiment Transfer for Humanoid Whole-Body Tracking
Any2Any transfers humanoid whole-body tracking models across embodiments via kinematic alignment followed by targeted PEFT, matching full-training performance with 1% of the data and compute on tested platforms.
-
What Drives Success in Physical Planning with Joint-Embedding Predictive World Models?
An empirical study of JEPA world models identifies architecture, training objective, and planning choices that yield a model outperforming DINO-WM and V-JEPA-2-AC on navigation and manipulation tasks.
-
MOBIUS: A Multi-Modal Bipedal Robot that can Walk, Crawl, Climb, and Roll
MOBIUS is a multi-modal bipedal robot with hybrid reinforcement learning and force control plus an MIQCP planner that enables walking, crawling, climbing, and rolling on varied terrains.