# BIND: Binding 3D Robot Actions to 2D Image Features

arXiv: https://arxiv.org/abs/2609.38443 · HTML: https://mlsanity.com/paper/2609.38443/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 33 of its 36 references that are in the map.
- First posted on arXiv 29 Sep 2026.

## Builds on (most cited first, 30 of 33)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Diffusion Policy: Visuomotor Policy Learning via Action Diffusion](https://mlsanity.com/paper/2303.04137/) (2023)
- [π0: A Vision-Language-Action Flow Model for General Robot Control](https://mlsanity.com/paper/2410.24164/) (2024)
- [Learning Fine-Grained Bimanual Manipulation with Low-Cost Hardware](https://mlsanity.com/paper/2304.13705/) (2023)
- [DINOv3](https://mlsanity.com/paper/2508.10104/) (2025)
- [LIBERO: Benchmarking Knowledge Transfer for Lifelong Robot Learning](https://mlsanity.com/paper/2306.03310/) (2023)
- [End-to-End Training of Deep Visuomotor Policies](https://mlsanity.com/paper/1504.00702/) (2015)
- [PaliGemma: A versatile 3B VLM for transfer](https://mlsanity.com/paper/2407.07726/) (2024)
- [Depth Anything 3: Recovering the Visual Space from Any Views](https://mlsanity.com/paper/2511.10647/) (2025)
- [CLIPort: What and Where Pathways for Robotic Manipulation](https://mlsanity.com/paper/2109.12098/) (2021)
- [Perceiver-Actor: A Multi-Task Transformer for Robotic Manipulation](https://mlsanity.com/paper/2209.05451/) (2022)
- [Learning Universal Policies via Text-Guided Video Generation](https://mlsanity.com/paper/2302.00111/) (2023)
- [Behavioral Cloning from Observation](https://mlsanity.com/paper/1805.01954/) (2018)
- [World Action Models are Zero-shot Policies](https://mlsanity.com/paper/2602.15922/) (2026)
- [Universal Manipulation Interface: In-The-Wild Robot Teaching Without In-The-Wild Robots](https://mlsanity.com/paper/2402.10329/) (2024)
- [Transporter Networks: Rearranging the Visual World for Robotic Manipulation](https://mlsanity.com/paper/2010.14406/) (2020)
- [Cosmos Policy: Fine-Tuning Video Models for Visuomotor Control and Planning](https://mlsanity.com/paper/2601.16163/) (2026)
- [One-Shot Imitation Learning](https://mlsanity.com/paper/1703.07326/) (2017)
- [RVT: Robotic View Transformer for 3D Object Manipulation](https://mlsanity.com/paper/2306.14896/) (2023)
- [Unified Video Action Model](https://mlsanity.com/paper/2503.00200/) (2025)
- [GELLO: A General, Low-Cost, and Intuitive Teleoperation Framework for Robot Manipulators](https://mlsanity.com/paper/2309.13037/) (2023)
- [RVT-2: Learning Precise Manipulation from Few Demonstrations](https://mlsanity.com/paper/2406.08545/) (2024)
- [Coarse-to-Fine Q-attention: Efficient Learning for Visual Robotic Manipulation via Discretisation](https://mlsanity.com/paper/2106.12534/) (2021)
- [RoboTwin: Dual-Arm Robot Benchmark with Generative Digital Twins](https://mlsanity.com/paper/2504.13059/) (2025)
- [GNFactor: Multi-Task Real Robot Learning with Generalizable Neural Feature Fields](https://mlsanity.com/paper/2308.16891/) (2023)
- [Motion Tracks: A Unified Representation for Human-Robot Transfer in Few-Shot Imitation Learning](https://mlsanity.com/paper/2501.06994/) (2025)
- [RAM: Retrieval-Based Affordance Transfer for Generalizable Zero-Shot Robotic Manipulation](https://mlsanity.com/paper/2407.04689/) (2024)
- [MolmoAct2: Action Reasoning Models for Real-world Deployment](https://mlsanity.com/paper/2605.02881/) (2026)
- [ManiFlow: A General Robot Manipulation Policy via Consistency Flow Training](https://mlsanity.com/paper/2509.01819/) (2025)
- [RT-Affordance: Affordances are Versatile Intermediate Representations for Robot Manipulation](https://mlsanity.com/paper/2411.02704/) (2024)

## Abstract (arXiv)

We introduce BIND, a new action representation for visuomotor robot policies that binds 3D robot actions to their corresponding 2D image features, yielding strong data efficiency gains and robustness to out-of-distribution object positions and camera viewpoints. The action heads of current robot policies are typically formulated as an MLP regression from a single global feature vector produced by a pre-trained vision encoder. This global formulation requires the policy network to discover, from demonstrations alone, the relationship between target robot actions and the image features they project onto. The consequence is that although modern image features are semantically descriptive, spatially robust, and even multiview-consistent, the policies built on them are brittle to subtle changes in camera viewpoint and object placement--and surprisingly data-inefficient. BIND closes this gap by supplying the action-feature relationship through camera geometry rather than learning: it discretizes a volume of candidate end effector positions, attaches each candidate to the pre-trained features at its projection in each camera view, and selects actions by scoring each candidate's position and image-bound feature combination. On a real robot, we study data efficiency and out-of-distribution robustness to unseen object positions and camera viewpoints, as well as general long-horizon task execution and dexterity. We find BIND to be highly data-efficient and robust: it achieves near-perfect success on tasks with as few as 5 demonstrations, and degrades gracefully under steep camera-viewpoint shifts and held-out object positions where coordinate-regression baselines completely fail.