# Vision-Based Manipulators Need to Also See from Their Hands

arXiv: https://arxiv.org/abs/2203.12677 · HTML: https://mlsanity.com/paper/2203.12677/

## Key facts

- Cited by 47 papers in the map (data through 14 Oct 2026).
- Builds on 43 of its 57 references that are in the map.
- First posted on arXiv 15 Mar 2022.

## Builds on (most cited first, 30 of 43)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actor](https://mlsanity.com/paper/1801.01290/) (2018)
- [Continuous control with deep reinforcement learning](https://mlsanity.com/paper/1509.02971/) (2015)
- [Playing Atari with Deep Reinforcement Learning](https://mlsanity.com/paper/1312.5602/) (2013)
- [Addressing Function Approximation Error in Actor-Critic Methods](https://mlsanity.com/paper/1802.09477/) (2018)
- [Describing Textures in the Wild](https://mlsanity.com/paper/1311.3618/) (2013)
- [Generative Adversarial Imitation Learning](https://mlsanity.com/paper/1606.03476/) (2016)
- [Distributionally Robust Neural Networks for Group Shifts: On the Importance of Regularization for Worst-Case Generalization](https://mlsanity.com/paper/1911.08731/) (2019)
- [Meta-World: A Benchmark and Evaluation for Multi-Task and Meta Reinforcement Learning](https://mlsanity.com/paper/1910.10897/) (2019)
- [Deep Variational Information Bottleneck](https://mlsanity.com/paper/1612.00410/) (2016)
- [Reconciling modern machine learning practice and the bias-variance trade-off](https://mlsanity.com/paper/1812.11118/) (2018)
- [End-to-End Training of Deep Visuomotor Policies](https://mlsanity.com/paper/1504.00702/) (2015)
- [QT-Opt: Scalable Deep Reinforcement Learning for Vision-Based Robotic Manipulation](https://mlsanity.com/paper/1806.10293/) (2018)
- [What Matters in Learning from Offline Human Demonstrations for Robot Manipulation](https://mlsanity.com/paper/2108.03298/) (2021)
- [Image Augmentation Is All You Need: Regularizing Deep Reinforcement Learning from Pixels](https://mlsanity.com/paper/2004.13649/) (2020)
- [Habitat 2.0: Training Home Assistants to Rearrange their Habitat](https://mlsanity.com/paper/2106.14405/) (2021)
- [Reinforcement Learning with Augmented Data](https://mlsanity.com/paper/2004.14990/) (2020)
- [Mastering Visual Continuous Control: Improved Data-Augmented Reinforcement Learning](https://mlsanity.com/paper/2107.09645/) (2021)
- [The Effectiveness of Data Augmentation in Image Classification using Deep Learning](https://mlsanity.com/paper/1712.04621/) (2017)
- [In Search of the Real Inductive Bias: On the Role of Implicit Regularization in Deep Learning](https://mlsanity.com/paper/1412.6614/) (2014)
- [Visual Foresight: Model-Based Deep Reinforcement Learning for Vision-Based Robotic Control](https://mlsanity.com/paper/1812.00568/) (2018)
- [One-Shot Visual Imitation Learning via Meta-Learning](https://mlsanity.com/paper/1709.04905/) (2017)
- [Discriminator-Actor-Critic: Addressing Sample Inefficiency and Reward Bias in Adversarial Imitation Learning](https://mlsanity.com/paper/1809.02925/) (2018)
- [Learning to Poke by Poking: Experiential Learning of Intuitive Physics](https://mlsanity.com/paper/1606.07419/) (2016)
- [Deep Spatial Autoencoders for Visuomotor Learning](https://mlsanity.com/paper/1509.06113/) (2015)
- [Reinforcement and Imitation Learning for Diverse Visuomotor Skills](https://mlsanity.com/paper/1802.09564/) (2018)
- [Grasping in the Wild:Learning 6DoF Closed-Loop Grasping from Low-Cost Demonstrations](https://mlsanity.com/paper/1912.04344/) (2019)
- [Variational Discriminator Bottleneck: Improving Imitation Learning, Inverse RL, and GANs by Constraining Information Flow](https://mlsanity.com/paper/1810.00821/) (2018)
- [Generalization in Reinforcement Learning with Selective Noise Injection and Information Bottleneck](https://mlsanity.com/paper/1910.12911/) (2019)
- [Offline Reinforcement Learning from Images with Latent Space Models](https://mlsanity.com/paper/2012.11547/) (2020)

## Built on by (most cited first, 30 of 47)

- [DayDreamer: World Models for Physical Robot Learning](https://mlsanity.com/paper/2206.14176/) (2022)
- [Mobile ALOHA: Learning Bimanual Mobile Manipulation with Low-Cost Whole-Body Teleoperation](https://mlsanity.com/paper/2401.02117/) (2024)
- [VIOLA: Imitation Learning for Vision-Based Manipulation with Object Proposal Priors](https://mlsanity.com/paper/2210.11339/) (2022)
- [Multi-View Masked World Models for Visual Robotic Manipulation](https://mlsanity.com/paper/2302.02408/) (2023)
- [Imitating Task and Motion Planning with Visuomotor Transformers](https://mlsanity.com/paper/2305.16309/) (2023)
- [Imitation Bootstrapped Reinforcement Learning](https://mlsanity.com/paper/2311.02198/) (2023)
- [Learning Generalizable Manipulation Policies with Object-Centric 3D Representations](https://mlsanity.com/paper/2310.14386/) (2023)
- [Multi-Stage Cable Routing through Hierarchical Imitation Learning](https://mlsanity.com/paper/2307.08927/) (2023)
- [NeRF in the Palm of Your Hand: Corrective Augmentation for Robotics via Novel-View Synthesis](https://mlsanity.com/paper/2301.08556/) (2023)
- [Surgical Robot Transformer (SRT): Imitation Learning for Surgical Tasks](https://mlsanity.com/paper/2407.12998/) (2024)
- [JUICER: Data-Efficient Imitation Learning for Robotic Assembly](https://mlsanity.com/paper/2404.03729/) (2024)
- [View-Invariant Policy Learning via Zero-Shot Novel View Synthesis](https://mlsanity.com/paper/2409.03685/) (2024)
- [Self-Improving Robots: End-to-End Autonomous Visuomotor Reinforcement Learning](https://mlsanity.com/paper/2303.01488/) (2023)
- [Frame Mining: a Free Lunch for Learning Robotic Manipulation from 3D Point Clouds](https://mlsanity.com/paper/2210.07442/) (2022)
- [Polybot: Training One Policy Across Robots While Embracing Variability](https://mlsanity.com/paper/2307.03719/) (2023)
- [SRT-H: A Hierarchical Framework for Autonomous Surgery via Language Conditioned Imitation Learning](https://mlsanity.com/paper/2505.10251/) (2025)
- [Local Policies Enable Zero-shot Long-horizon Manipulation](https://mlsanity.com/paper/2410.22332/) (2024)
- [One-Shot Imitation Learning: A Pose Estimation Perspective](https://mlsanity.com/paper/2310.12077/) (2023)
- [AirExo-2: Scaling up Generalizable Robotic Imitation Learning with Low-Cost Exoskeletons](https://mlsanity.com/paper/2503.03081/) (2025)
- [Learning Viewpoint-Agnostic Visual Representations by Recovering Tokens in 3D Space](https://mlsanity.com/paper/2206.11895/) (2022)
- [The Treachery of Images: Bayesian Scene Keypoints for Deep Policy Learning in Robotic Manipulation](https://mlsanity.com/paper/2305.04718/) (2023)
- [D5RL: Diverse Datasets for Data-Driven Deep Reinforcement Learning](https://mlsanity.com/paper/2408.08441/) (2024)
- [Real-is-Sim: Bridging the Sim-to-Real Gap with a Dynamic Digital Twin](https://mlsanity.com/paper/2504.03597/) (2025)
- [Part-Guided 3D RL for Sim2Real Articulated Object Manipulation](https://mlsanity.com/paper/2404.17302/) (2024)
- [Multi-view Disentanglement for Reinforcement Learning with Multiple Cameras](https://mlsanity.com/paper/2404.14064/) (2024)
- [Bridging Perspectives: A Survey on Cross-view Collaborative Intelligence with Egocentric-Exocentric Vision](https://mlsanity.com/paper/2506.06253/) (2025)
- [Visual-Policy Learning through Multi-Camera View to Single-Camera View Knowledge Distillation for Robot Manipulation Tasks](https://mlsanity.com/paper/2303.07026/) (2023)
- [The Surprising Ineffectiveness of Pre-Trained Visual Representations for Model-Based Reinforcement Learning](https://mlsanity.com/paper/2411.10175/) (2024)
- [Giving Robots a Hand: Learning Generalizable Manipulation with Eye-in-Hand Human Video Demonstrations](https://mlsanity.com/paper/2307.05959/) (2023)
- [Egocentric and Exocentric Methods: A Short Survey](https://mlsanity.com/paper/2410.20621/) (2024)

## Abstract (arXiv)

We study how the choice of visual perspective affects learning and generalization in the context of physical manipulation from raw sensor observations. Compared with the more commonly used global third-person perspective, a hand-centric (eye-in-hand) perspective affords reduced observability, but we find that it consistently improves training efficiency and out-of-distribution generalization. These benefits hold across a variety of learning algorithms, experimental settings, and distribution shifts, and for both simulated and real robot apparatuses. However, this is only the case when hand-centric observability is sufficient; otherwise, including a third-person perspective is necessary for learning, but also harms out-of-distribution generalization. To mitigate this, we propose to regularize the third-person information stream via a variational information bottleneck. On six representative manipulation tasks with varying hand-centric observability adapted from the Meta-World benchmark, this results in a state-of-the-art reinforcement learning agent operating from both perspectives improving its out-of-distribution generalization on every task. While some practitioners have long put cameras in the hands of robots, our work systematically analyzes the benefits of doing so and provides simple and broadly applicable insights for improving end-to-end learned vision-based robotic manipulation.