ERMV: Editing 4D Robotic Multi-view images to enhance embodied agents

TL;DR

ERMV is a novel data augmentation framework for editing 4D multi-view robotic images, improving embodied agent training by maintaining consistency, expanding editing scope, and ensuring semantic integrity.

Contribution

The paper introduces ERMV, a new method for editing 4D robotic multi-view data that addresses key challenges in consistency, efficiency, and semantic preservation.

Findings

ERMV significantly improves model robustness in simulated environments.

ERMV enhances generalization of embodied intelligence policies in real-world tests.

The proposed framework reduces computational costs while maintaining high editing quality.

Abstract

Robot imitation learning relies on 4D multi-view sequential images. However, the high cost of data collection and the scarcity of high-quality data severely constrain the generalization and application of embodied intelligence policies like Vision-Language-Action (VLA) models. Data augmentation is a powerful strategy to overcome data scarcity, but methods for editing 4D multi-view sequential images for manipulation tasks are currently lacking. Thus, we propose ERMV (Editing Robotic Multi-View 4D data), a novel data augmentation framework that efficiently edits an entire multi-view sequence based on single-frame editing and robot state conditions. This task presents three core challenges: (1) maintaining geometric and appearance consistency across dynamic views and long time horizons; (2) expanding the working window with low computational costs; and (3) ensuring the semantic integrity of critical objects like the robot arm. ERMV addresses these challenges through a series of innovations. First, to ensure spatio-temporal consistency in motion blur, we introduce a novel Epipolar Motion-Aware Attention (EMA-Attn) mechanism that learns pixel shift caused by movement before applying geometric constraints. Second, to maximize the editing working window, ERMV pioneers a Sparse Spatio-Temporal (STT) module, which decouples the temporal and spatial views and remodels a single-frame multi-view problem through sparse sampling of the views to reduce computational demands. Third, to alleviate error accumulation, we incorporate a feedback intervention Mechanism, which uses a Multimodal Large Language Model (MLLM) to check editing inconsistencies and request targeted expert guidance only when necessary. Extensive experiments demonstrate that ERMV-augmented data significantly boosts the robustness and generalization of VLA models in both simulated and real-world environments.