Learning to Drive is a Free Gift: Large-Scale Label-Free Autonomy Pretraining from Unposed In-The-Wild Videos
This work addresses the challenge of scalable perception for autonomous driving by leveraging abundant unposed video data, offering a potential foundation model for the field.
The authors tackled the problem of learning autonomous driving representations from unlabeled in-the-wild videos by proposing a label-free, teacher-guided framework that jointly predicts point maps, poses, segmentation, and motion masks, achieving state-of-the-art performance on the NAVSIM benchmark with a single monocular camera.
Ego-centric driving videos available online provide an abundant source of visual data for autonomous driving, yet their lack of annotations makes it difficult to learn representations that capture both semantic structure and 3D geometry. Recent advances in large feedforward spatial models demonstrate that point maps and ego-motion can be inferred in a single forward pass, suggesting a promising direction for scalable driving perception. We therefore propose a label-free, teacher-guided framework for learning autonomous driving representations directly from unposed videos. Unlike prior self-supervised approaches that focus primarily on frame-to-frame consistency, we posit that safe and reactive driving depends critically on temporal context. To this end, we leverage a feedforward architecture equipped with a lightweight autoregressive module, trained using multi-modal supervisory signals that guide the model to jointly predict current and future point maps, camera poses, semantic segmentation, and motion masks. Multi-modal teachers provide sequence-level pseudo-supervision, enabling LFG to learn a unified pseudo-4D representation from raw YouTube videos without poses, labels, or LiDAR. The resulting encoder not only transfers effectively to downstream autonomous driving planning on the NAVSIM benchmark, surpassing multi-camera and LiDAR baselines with only a single monocular camera, but also yields strong performance when evaluated on a range of semantic, geometric, and qualitative motion prediction tasks. These geometry and motion-aware features position LFG as a compelling video-centric foundation model for autonomous driving.