Natural Language Can Help Bridge the Sim2Real Gap

• URL: http://arxiv.org/abs/2405.10020v2
• Date: Tue, 2 Jul 2024 07:29:04 GMT
• Title: Natural Language Can Help Bridge the Sim2Real Gap
• Authors: Albert Yu, Adeline Foote, Raymond Mooney, Roberto Martín-Martín,
• Abstract summary: Sim2Real is a promising paradigm for overcoming data scarcity in the real-world target domain. We propose using natural language descriptions of images as a unifying signal across domains. We demonstrate that training the image encoder to predict the language description serves as a useful, data-efficient pretraining step.
• Score: 9.458180590551715
• License: http://creativecommons.org/licenses/by-sa/4.0/
• Abstract: The main challenge in learning image-conditioned robotic policies is acquiring a visual representation conducive to low-level control. Due to the high dimensionality of the image space, learning a good visual representation requires a considerable amount of visual data. However, when learning in the real world, data is expensive. Sim2Real is a promising paradigm for overcoming data scarcity in the real-world target domain by using a simulator to collect large amounts of cheap data closely related to the target task. However, it is difficult to transfer an image-conditioned policy from sim to real when the domains are very visually dissimilar. To bridge the sim2real visual gap, we propose using natural language descriptions of images as a unifying signal across domains that captures the underlying task-relevant semantics. Our key insight is that if two image observations from different domains are labeled with similar language, the policy should predict similar action distributions for both images. We demonstrate that training the image encoder to predict the language description or the distance between descriptions of a sim or real image serves as a useful, data-efficient pretraining step that helps learn a domain-invariant image representation. We can then use this image encoder as the backbone of an IL policy trained simultaneously on a large amount of simulated and a handful of real demonstrations. Our approach outperforms widely used prior sim2real methods and strong vision-language pretraining baselines like CLIP and R3M by 25 to 40%. See additional videos and materials at https://robin-lab.cs.utexas.edu/lang4sim2real/.

Related papers

• Towards Pragmatic Semantic Image Synthesis for Urban Scenes [4.36080478413575]
  We present a new task: given a dataset with synthetic images and labels and a dataset with unlabeled real images, our goal is to learn a model that can generate images with the content of the input mask and the appearance of real images. We leverage the synthetic image as a guide to the content of the generated image by penalizing the difference between their high-level features on a patch level. In contrast to previous works which employ one discriminator that overfits the target domain semantic distribution, we employ a discriminator for the whole image and multiscale discriminators on the image patches.
  arXiv  Detail & Related papers  (2023-05-16T18:01:12Z)
• CLIP$^2$: Contrastive Language-Image-Point Pretraining from Real-World Point Cloud Data [80.42480679542697]
  We propose Contrastive Language-Image-Point Cloud Pretraining (CLIP$2$) to learn the transferable 3D point cloud representation in realistic scenarios. Specifically, we exploit naturally-existed correspondences in 2D and 3D scenarios, and build well-aligned and instance-based text-image-point proxies from those complex scenarios.
  arXiv  Detail & Related papers  (2023-03-22T09:32:45Z)
• AptSim2Real: Approximately-Paired Sim-to-Real Image Translation [8.208569626646035]
  Sim-to-real transfer modifies simulated images to better match real-world data. AptSim2Real exploits the fact that simulators can generate scenes loosely resembling real-world scenes in terms of lighting, environment, and composition. Our novel training strategy results in significant qualitative and quantitative improvements, with up to a 24% improvement in FID score.
  arXiv  Detail & Related papers  (2023-03-09T06:18:44Z)
• Sim2real Transfer Learning for Point Cloud Segmentation: An Industrial Application Case on Autonomous Disassembly [55.41644538483948]
  We present an industrial application case that uses sim2real transfer learning for point cloud data. We provide insights on how to generate and process synthetic point cloud data. A novel patch-based attention network is proposed additionally to tackle this problem.
  arXiv  Detail & Related papers  (2023-01-12T14:00:37Z)
• A Shared Representation for Photorealistic Driving Simulators [83.5985178314263]
  We propose to improve the quality of generated images by rethinking the discriminator architecture. The focus is on the class of problems where images are generated given semantic inputs, such as scene segmentation maps or human body poses. We aim to learn a shared latent representation that encodes enough information to jointly do semantic segmentation, content reconstruction, along with a coarse-to-fine grained adversarial reasoning.
  arXiv  Detail & Related papers  (2021-12-09T18:59:21Z)
• Align before Fuse: Vision and Language Representation Learning with Momentum Distillation [52.40490994871753]
  We introduce a contrastive loss to representations BEfore Fusing (ALBEF) through cross-modal attention. We propose momentum distillation, a self-training method which learns from pseudo-targets produced by a momentum model. ALBEF achieves state-of-the-art performance on multiple downstream vision-language tasks.
  arXiv  Detail & Related papers  (2021-07-16T00:19:22Z)
• Curious Representation Learning for Embodied Intelligence [81.21764276106924]
  Self-supervised representation learning has achieved remarkable success in recent years. Yet to build truly intelligent agents, we must construct representation learning algorithms that can learn from environments. We propose a framework, curious representation learning, which jointly learns a reinforcement learning policy and a visual representation model.
  arXiv  Detail & Related papers  (2021-05-03T17:59:20Z)
• Scaling Up Visual and Vision-Language Representation Learning With Noisy Text Supervision [57.031588264841]
  We leverage a noisy dataset of over one billion image alt-text pairs, obtained without expensive filtering or post-processing steps. A simple dual-encoder architecture learns to align visual and language representations of the image and text pairs using a contrastive loss. We show that the scale of our corpus can make up for its noise and leads to state-of-the-art representations even with such a simple learning scheme.
  arXiv  Detail & Related papers  (2021-02-11T10:08:12Z)
• Action Image Representation: Learning Scalable Deep Grasping Policies with Zero Real World Data [12.554739620645917]
  Action Image represents a grasp proposal as an image and uses a deep convolutional network to infer grasp quality. We show that this representation works on a variety of inputs, including color images (RGB), depth images (D) and combined color-depth (RGB-D)
  arXiv  Detail & Related papers  (2020-05-13T21:40:21Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.