Pith. sign in

REVIEW 20 cited by

Novel Demonstration Generation with Gaussian Splatting Enables Robust One-Shot Manipulation

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2504.13175 v1 pith:FCMAJIMS submitted 2025-04-17 cs.RO

Novel Demonstration Generation with Gaussian Splatting Enables Robust One-Shot Manipulation

classification cs.RO
keywords datatypesdemonstrationsdiversegaussiangeneralizationnovelobject
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
read the original abstract

Visuomotor policies learned from teleoperated demonstrations face challenges such as lengthy data collection, high costs, and limited data diversity. Existing approaches address these issues by augmenting image observations in RGB space or employing Real-to-Sim-to-Real pipelines based on physical simulators. However, the former is constrained to 2D data augmentation, while the latter suffers from imprecise physical simulation caused by inaccurate geometric reconstruction. This paper introduces RoboSplat, a novel method that generates diverse, visually realistic demonstrations by directly manipulating 3D Gaussians. Specifically, we reconstruct the scene through 3D Gaussian Splatting (3DGS), directly edit the reconstructed scene, and augment data across six types of generalization with five techniques: 3D Gaussian replacement for varying object types, scene appearance, and robot embodiments; equivariant transformations for different object poses; visual attribute editing for various lighting conditions; novel view synthesis for new camera perspectives; and 3D content generation for diverse object types. Comprehensive real-world experiments demonstrate that RoboSplat significantly enhances the generalization of visuomotor policies under diverse disturbances. Notably, while policies trained on hundreds of real-world demonstrations with additional 2D data augmentation achieve an average success rate of 57.2%, RoboSplat attains 87.8% in one-shot settings across six types of generalization in the real world.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 20 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. DeformGen: Dynamics-Based Topology Augmentation for Deformable Manipulation Policy Learning

    cs.RO 2026-06 unverdicted novelty 7.0

    DeformGen uses dynamics-based state expansion via localized disturbances and deformation-field warping for trajectory transfer to improve policy learning on deformable manipulation benchmarks.

  2. 3D Generation for Embodied AI and Robotic Simulation: A Survey

    cs.RO 2026-04 accept novelty 7.0

    3D generation for embodied AI is shifting from visual realism toward interaction readiness, organized into data generation, simulation environments, and sim-to-real bridging roles.

  3. RoboSnap: One-Shot Real-to-Sim Scene Generation for Generalizable Robot Learning and Evaluation

    cs.RO 2026-07 conditional novelty 6.0

    A single RGB image is converted into a layered, simulation-ready robot scene that supports trajectory replay, synthetic data generation, and meaningful sim-real policy evaluation, plus a 564-scene DROID-Sim companion set.

  4. FalconTrack: Photorealistic Auto-Labeled Perception and Physics-Aware Vision-Based Aerial Tracking

    cs.RO 2026-06 unverdicted novelty 6.0

    FalconTrack automates photorealistic dataset creation via Gaussian Splatting and achieves high zero-shot sim-to-real performance in vision-based aerial tracking using multi-head perception and class-conditioned EKF.

  5. DeformGen: Dynamics-Based Topology Augmentation for Deformable Manipulation Policy Learning

    cs.RO 2026-06 unverdicted novelty 6.0

    DeformGen augments states via localized disturbance simulation and trajectories via deformation-field warping to improve deformable manipulation policy learning over original data and rigid baselines.

  6. UniviewVLA: A Unified Multiview Vision-Language-Action Model with World Modeling

    cs.RO 2026-06 unverdicted novelty 6.0

    UniviewVLA generates multiview future views from two cameras via world modeling, plus token compression and view selection, to boost occlusion handling in robot manipulation while matching standard benchmark performance.

  7. One Demo is Worth a Thousand Trajectories: Action-View Augmentation for Visuomotor Policies

    cs.RO 2026-06 unverdicted novelty 6.0

    A framework augments single fisheye demonstrations into multiple novel-view trajectories with obstacles via fisheye-adapted Gaussian Splatting and trajectory optimization, raising policy success rates in original and ...

  8. An Efficient Metric for Data Quality Measurement in Imitation Learning

    cs.RO 2026-05 unverdicted novelty 6.0

    Power spectral density of trajectories ranks demonstration quality for imitation learning, enabling rollout-free curation that improves fine-tuned policy success.

  9. A Principled Approach for Creating High-fidelity Synthetic Demonstrations for Imitation Learning

    cs.RO 2026-05 unverdicted novelty 6.0

    DMP retargeting within 3DGS scenes preserves expert motion shape and phase to create diverse yet high-fidelity demonstrations, yielding lower deviation, fewer collisions, and higher downstream policy success than plan...

  10. GS-Playground: A High-Throughput Photorealistic Simulator for Vision-Informed Robot Learning

    cs.RO 2026-04 unverdicted novelty 6.0

    GS-Playground delivers a high-throughput photorealistic simulator for vision-informed robot learning via parallel physics integrated with batch 3D Gaussian Splatting at 10^4 FPS and an automated Real2Sim workflow for ...

  11. ShapeGen: Robotic Data Generation for Category-Level Manipulation

    cs.RO 2026-04 unverdicted novelty 6.0

    ShapeGen generates shape-diverse 3D robotic manipulation demonstrations without simulators by curating a functional shape library and applying a minimal-annotation pipeline for novel, physically plausible data.

  12. WARPED: Wrist-Aligned Rendering for Robot Policy Learning from Egocentric Human Demonstrations

    cs.RO 2026-04 unverdicted novelty 6.0

    WARPED synthesizes realistic wrist-view observations from monocular egocentric human videos via foundation models, hand-object tracking, retargeting, and Gaussian Splatting to train visuomotor policies that match tele...

  13. SIM1: Physics-Aligned Simulator as Zero-Shot Data Scaler in Deformable Worlds

    cs.RO 2026-04 unverdicted novelty 6.0

    SIM1 converts sparse real demonstrations into high-fidelity synthetic data through physics-aligned simulation, yielding policies that match real-data performance at a 1:15 ratio with 90% zero-shot success on deformabl...

  14. TwinRL: Digital Twin-Driven Reinforcement Learning for Real-World Robotic Manipulation

    cs.RO 2026-02 unverdicted novelty 6.0

    TwinRL expands RL exploration via digital twin reconstruction and twin RL warm-up to guide real-world learning, reaching near-100% success with 20 minutes of on-robot time across four tasks.

  15. GraspVLA: a Grasping Foundation Model Pre-trained on Billion-scale Synthetic Action Data

    cs.RO 2025-05 unverdicted novelty 6.0

    GraspVLA shows that pretraining a grasping model on a billion synthetic action frames enables zero-shot open-vocabulary performance and sim-to-real transfer.

  16. FrozenDrive: Zero-Shot Text-Guided Driving Scene Generation and Data Augmentation with Parameter-Free Frozen Diffusion Model

    cs.CV 2026-06 unverdicted novelty 5.0

    FrozenDrive enables zero-shot text-guided generation of consistent multi-view driving scenes via a parameter-free frozen diffusion backbone with spatio-temporal attention, improving autonomous driving models on advers...

  17. ManiSplat: Manipulation Trajectory Synthesis from Monocular Video via Decoupled 3D Gaussian Splatting

    cs.CV 2026-06 unverdicted novelty 5.0

    ManiSplat introduces a graph-structured disentangled 3D Gaussian framework with task-oriented alignment to reconstruct controllable dynamic scenes from monocular ego-view robotic videos.

  18. HyperSim: A Holistic Sim-To-Real Framework For Robust Robotic Manipulation

    cs.RO 2026-05 unverdicted novelty 5.0

    HyperSim reports 80% and 95% sim-to-real success on two manipulation policies across 400 real executions by combining synthetic environment synthesis, adversarial trajectories, and co-training.

  19. 3D Generation for Embodied AI and Robotic Simulation: A Survey

    cs.RO 2026-04 unverdicted novelty 3.0

    The survey organizes 3D generation for embodied AI into data generators for assets, simulation environments for interaction, and sim-to-real bridges, noting a shift toward interaction readiness and listing bottlenecks...

  20. 3D Generation for Embodied AI and Robotic Simulation: A Survey

    cs.RO 2026-04 unverdicted novelty 2.0

    The paper surveys 3D generation techniques for embodied AI and robotics, categorizing them into data generation, simulation environments, and sim-to-real bridging while identifying bottlenecks in physical validity and...