Controllable Pedestrian Video Editing for Multi-View Driving Scenarios via Motion Sequence

TL;DR

This paper introduces a novel framework for controllable pedestrian video editing in multi-view driving scenarios, enhancing data augmentation and scenario simulation for autonomous driving systems.

Contribution

It presents a new method combining video inpainting and human motion control to enable flexible, multi-view pedestrian editing with high realism and consistency.

Findings

Achieves high-quality pedestrian editing with visual realism

Maintains spatiotemporal coherence across views

Demonstrates robustness and versatility in multi-view scenarios

Abstract

Pedestrian detection models in autonomous driving systems often lack robustness due to insufficient representation of dangerous pedestrian scenarios in training datasets. To address this limitation, we present a novel framework for controllable pedestrian video editing in multi-view driving scenarios by integrating video inpainting and human motion control techniques. Our approach begins by identifying pedestrian regions of interest across multiple camera views, expanding detection bounding boxes with a fixed ratio, and resizing and stitching these regions into a unified canvas while preserving cross-view spatial relationships. A binary mask is then applied to designate the editable area, within which pedestrian editing is guided by pose sequence control conditions. This enables flexible editing functionalities, including pedestrian insertion, replacement, and removal. Extensive experiments demonstrate that our framework achieves high-quality pedestrian editing with strong visual realism, spatiotemporal coherence, and cross-view consistency. These results establish the proposed method as a robust and versatile solution for multi-view pedestrian video generation, with broad potential for applications in data augmentation and scenario simulation in autonomous driving.