AimBot: A Simple Auxiliary Visual Cue to Enhance Spatial Awareness of Visuomotor Policies

2025/08/11 by Yinpei Dai, Dai, Yinpei, Jayjun Lee +14 · 1 citation
Computer Science · Engineering · Neuroscience · #Advanced Vision and Imaging #FOS: Computer and information sciences #Motor Control and Adaptation #Robot Manipulation and Learning #Robotics (cs.RO)

paper · pdf · doi:10.48550/arxiv.2508.08113

openalex publication_date 2025/08/11 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

In this paper, we propose AimBot, a lightweight visual augmentation technique that provides explicit spatial cues to improve visuomotor policy learning in robotic manipulation. AimBot overlays shooting lines and scope reticles onto multi-view RGB images, offering auxiliary visual guidance that encodes the end-effector's state. The overlays are computed from depth images, camera extrinsics, and the current end-effector pose, explicitly conveying spatial relationships between the gripper and objects in the scene. AimBot incurs minimal computational overhead (less than 1 ms) and requires no changes to model architectures, as it simply replaces original RGB images with augmented counterparts. Despite its simplicity, our results show that AimBot consistently improves the performance of various visuomotor policies in both simulation and real-world settings, highlighting the benefits of spatially grounded visual feedback.

Citations

• RoboRefer: Towards Spatial Referring with Reasoning in Vision-Language Models for Robotics
• OneTwoVLA: A Unified Vision-Language-Action Model with Adaptive Reasoning
• NORA: A Small Open-Sourced Generalist Vision Language Action Model for Embodied Tasks
• GR00T N1: An Open Foundation Model for Generalist Humanoid Robots
• PointVLA: Injecting the 3D World into Vision-Language-Action Models
• AgiBot World Colosseo: A Large-scale Manipulation Platform for Scalable and Intelligent Embodied Systems
• Fine-Tuning Vision-Language-Action Models: Optimizing Speed and Success
• SpatialVLA: Exploring Spatial Representations for Visual-Language-Action Model
• FAST: Efficient Action Tokenization for Vision-Language-Action Models
• Beyond Sight: Finetuning Generalist Robot Policies with Heterogeneous Sensors via Language Grounding
• TraceVLA: Visual Trace Prompting Enhances Spatial-Temporal Awareness for Generalist Robotic Policies
• RoboSpatial: Teaching Spatial Understanding to 2D and 3D Vision-Language Models for Robotics
• RT-Affordance: Affordances are Versatile Intermediate Representations for Robot Manipulation
• π₀: A Vision-Language-Action Flow Model for General Robot Control
• Teaching Embodied Reinforcement Learning Agents: Informativeness and Diversity of Language Use
• RACER: Rich Language-Guided Failure Recovery Policies for Imitation Learning
• Generative Image as Action Models
• RoboPoint: A Vision-Language Model for Spatial Affordance Prediction for Robotics
• OpenVLA: An Open-Source Vision-Language-Action Model
• DROID: A Large-Scale In-The-Wild Robot Manipulation Dataset
• Gemma: Open Models Based on Gemini Research and Technology
• Universal Manipulation Interface: In-The-Wild Robot Teaching Without In-The-Wild Robots
• PIVOT: Iterative Visual Prompting Elicits Actionable Knowledge for VLMs
• Foundation Models in Robotics: Applications, Challenges, and the Future
• Visual In-Context Prompting
• Think, Act, and Ask: Open-World Interactive Personalized Robot Navigation
• Qwen Technical Report
• Llama 2: Open Foundation and Fine-Tuned Chat Models
• LIBERO: Benchmarking Knowledge Transfer for Lifelong Robot Learning
• What does CLIP know about a red circle? Visual prompt engineering for VLMs
• GPT-4 Technical Report
• Diffusion Policy: Visuomotor Policy Learning via Action Diffusion
• Diffusion policy: Visuomotor policy learning via action diffusion
• GALAXY: A Generative Pre-trained Model for Task-Oriented Dialog with Semi-Supervised Learning and Explicit Policy Injection
• End-to-End Training of Deep Visuomotor Policies

Cited by

• SutureBot: A Precision Framework & Benchmark For Autonomous End-to-End Suturing

Related

• Do You Need Proprioceptive States in Visuomotor Policies?
• ImageManip: Image-based Robotic Manipulation with Affordance-guided Next View Selection
• Goal-Conditioned End-to-End Visuomotor Control for Versatile Skill Primitives
• Look, Zoom, Understand: The Robotic Eyeball for Embodied Perception
• VGGT-DP: Generalizable Robot Control via Vision Foundation Models