OpenLongTail: Generative Scaling of Long-Tail Driving Data

2026-07-10

AI Summary

This work addresses the scarcity of long-tail driving scenarios, which are typically captured as monocular, heterogeneous videos lacking multi-view coverage, thereby limiting the robustness of autonomous driving policies. To overcome this, the authors propose the first open-source generative data engine that transforms real-world monocular long-tail videos into view-aligned, temporally coherent multi-view training data. The method leverages pose-guided extrapolated view synthesis combined with Plücker ray-based geometric modeling to achieve high cross-view and temporal consistency. Experimental results demonstrate that the generated data significantly enhances closed-loop driving robustness under long-tail events and validates its effectiveness through metrics including view fidelity, multi-view consistency, and ego-trajectory recovery accuracy.

Abstract

Scaling robust driving policies is fundamentally bottlenecked by the scarcity of edge cases in curated datasets. While the real world continuously captures these critical events, such long-tail events remain underutilized when collected from heterogeneous sources. Specifically, diverse but valuable in-the-wild long-tail videos lack the full view coverage required for training policy models, often missing multi-view poses or originating solely from monocular dash cameras. This modality gap prevents these ubiquitous observations from being converted into scalable training data for long-tail generalization. We introduce OpenLongTail, an open-source generative data engine for scaling autonomous driving policies under long-tail events. To transform heterogeneous data sources into view-aligned and temporally coherent multi-view assets that are useful for policy learning, we develop a pose-informed extrapolative view synthesis pipeline that generates the missing views. We further enhance cross-view consistency and the temporal alignment for the newly generated views by injecting Plücker ray geometry into the scalable generation engine. By synthesizing heterogeneous long-tail data, we observe a significant improvement in closed-loop driving robustness in handling long-tail events. By measuring the extrapolative view synthesis and pose metrics, we validate the effectiveness of OpenLongTail in visual fidelity, cross-view consistency, and ego-trajectory recovery.

Problem

• long-tail driving
• autonomous driving
• multi-view synthesis
• edge cases
• heterogeneous data

Innovation

• generative data engine
• long-tail driving
• view synthesis
• Plücker ray geometry
• multi-view consistency