Learning to Augment Synthetic Images for Sim2Real Policy Transfer

Alexander Pashevich 1 Robin Strudel 2 Igor Kalevatykh 2 Ivan Laptev 2 Cordelia Schmid 1
1 Thoth - Apprentissage de modèles à partir de données massives
Inria Grenoble - Rhône-Alpes, LJK - Laboratoire Jean Kuntzmann
2 WILLOW - Models of visual object recognition and scene understanding
Inria de Paris, DI-ENS - Département d'informatique de l'École normale supérieure
Abstract : Vision and learning have made significant progress that could improve robotics policies for complex tasks and environments. Learning deep neural networks for image understanding, however, requires large amounts of domain-specific visual data. While collecting such data from real robots is possible, such an approach limits the scalability as learning policies typically requires thousands of trials. In this work we attempt to learn manipulation policies in simulated environments. Simulators enable scalability and provide access to the underlying world state during training. Policies learned in simulators, however, do not transfer well to real scenes given the domain gap between real and synthetic data. We follow recent work on domain randomization and augment synthetic images with sequences of random transformations. Our main contribution is to optimize the augmentation strategy for sim2real transfer and to enable domain-independent policy learning. We design an efficient search for depth image augmentations using object localization as a proxy task. Given the resulting sequence of random transformations, we use it to augment synthetic depth images during policy learning. Our augmentation strategy is policy-independent and enables policy learning with no real images. We demonstrate our approach to significantly improve accuracy on three manipulation tasks evaluated on a real robot.
Complete list of metadatas

https://hal.archives-ouvertes.fr/hal-02273326
Contributor : Alexander Pashevich <>
Submitted on : Wednesday, August 28, 2019 - 5:29:12 PM
Last modification on : Monday, September 2, 2019 - 11:28:04 AM

Links full text

Identifiers

  • HAL Id : hal-02273326, version 1
  • ARXIV : 1903.07740

Collections

Citation

Alexander Pashevich, Robin Strudel, Igor Kalevatykh, Ivan Laptev, Cordelia Schmid. Learning to Augment Synthetic Images for Sim2Real Policy Transfer. IROS 2019 - IEEE/RSJ International Conference on Intelligent Robots and Systems, Nov 2019, Macao, China. ⟨hal-02273326⟩

Share

Metrics

Record views

38