Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

ROPA: Synthetic Robot Pose Generation for RGB-D Bimanual Data Augmentation

About

Training robust bimanual manipulation policies via imitation learning requires demonstration data with broad coverage over robot poses, contacts, and scene contexts. However, collecting diverse and precise real-world demonstrations is costly and time-consuming, which hinders scalability. Prior works have addressed this with data augmentation, typically for either eye-in-hand (wrist camera) setups with RGB inputs or for generating novel images without paired actions, leaving augmentation for eye-to-hand (third-person) RGB-D training with new action labels less explored. In this paper, we propose Synthetic Robot Pose Generation for RGB-D Bimanual Data Augmentation (ROPA), an offline imitation learning data augmentation method that fine-tunes Stable Diffusion to synthesize third-person RGB and RGB-D observations of novel robot poses. Our approach simultaneously generates corresponding joint-space action labels while employing constrained optimization to enforce physical consistency through appropriate gripper-to-object contact constraints in bimanual scenarios. We evaluate our method on 5 simulated and 3 real-world tasks. Our results across 2625 simulation trials and 300 real-world trials demonstrate that ROPA outperforms baselines and ablations, showing its potential for scalable RGB and RGB-D data augmentation in eye-to-hand bimanual manipulation. Our project website is available at: https://ropaaug.github.io/.

Jason Chen, I-Chun Arthur Liu, Gaurav Sukhatme, Daniel Seita• 2025

Related benchmarks

TaskDatasetResultRank
Bimanual Robot ManipulationCoordinated Lift Ball (CLB) Simulation
Success Rate72
7
Bimanual Robot ManipulationCoordinated Lift Tray (CLT) Simulation
Success Rate35
7
Bimanual Robot ManipulationCoordinated Push Box (CPB) Simulation
Success Rate62.7
7
Bimanual Robot ManipulationBig Sweep Round (BSR) Simulation
Success Rate29
7
Bimanual Robot ManipulationCoordinated Pick In Drawer (CPID) Simulation
Success Rate35
7
Push BlockReal-world
Success Rate20
6
Lift BallReal-world
Success Rate95
5
Lift DrawerReal-world
Success Rate0.65
5
Showing 8 of 8 rows

Other info

Follow for update