GeoReFormer: Geometry-Aware Refinement for Lane Segment Detection and Topology Reasoning
This work addresses the challenge of accurate online map construction for autonomous driving by improving lane detection and topology reasoning, representing an incremental advance over existing transformer methods.
The paper tackled the problem of 3D lane segment detection and topology reasoning for autonomous driving map construction by proposing GeoReFormer, a transformer-based architecture that embeds geometry- and topology-aware inductive biases, achieving state-of-the-art performance with 34.5% mAP on the OpenLane-V2 benchmark.
Accurate 3D lane segment detection and topology reasoning are critical for structured online map construction in autonomous driving. Recent transformer-based approaches formulate this task as query-based set prediction, yet largely inherit decoder designs originally developed for compact object detection. However, lane segments are continuous polylines embedded in directed graphs, and generic query initialization and unconstrained refinement do not explicitly encode this geometric and relational structure. We propose GeoReFormer (Geometry-aware Refinement Transformer), a unified query-based architecture that embeds geometry- and topology-aware inductive biases directly within the transformer decoder. GeoReFormer introduces data-driven geometric priors for structured query initialization, bounded coordinate-space refinement for stable polyline deformation, and per-query gated topology propagation to selectively integrate relational context. On the OpenLane-V2 benchmark, GeoReFormer achieves state-of-the-art performance with 34.5% mAP while improving topology consistency over strong transformer baselines, demonstrating the utility of explicit geometric and relational structure encoding.