ROI-Driven Foveated Attention for Unified Egocentric Representations in Vision-Language-Action Systems

TL;DR

This paper introduces a ROI-driven egocentric data representation for embodied AI, improving cross-embodiment transfer and reducing data collection costs by focusing on contact-critical regions in vision-language-action systems.

Contribution

It proposes a geometry-grounded, hand-centric ROI pipeline that enhances data reuse and transferability across different robots in embodied AI systems.

Findings

ROI preserves high local information density in contact regions

Pipeline enables cross-embodiment data sharing and transfer

Improves scalability of vision-language-action models

Abstract

The development of embodied AI systems is increasingly constrained by the availability and structure of physical interaction data. Despite recent advances in vision-language-action (VLA) models, current pipelines suffer from high data collection cost, limited cross-embodiment alignment, and poor transfer from internet-scale visual data to robot control. We propose a region-of-interest (ROI) driven engineering workflow that introduces an egocentric, geometry-grounded data representation. By projecting end-effector poses via forward kinematics (FK) into a single external camera, we derive movement-aligned hand-centric ROIs without requiring wrist-mounted cameras or multi-view systems. Unlike directly downsampling the full frame, ROI is cropped from the original image before resizing, preserving high local information density for contact-critical regions while retaining global context. We present a reproducible pipeline covering calibration, synchronization, ROI generation, deterministic boundary handling, and metadata governance. The resulting representation is embodiment-aligned and viewpoint-normalized, enabling data reuse across heterogeneous robots. We argue that egocentric ROI serves as a practical data abstraction for scalable collection and cross-embodiment learning, bridging internet-scale perception and robot-specific control.