LaneGraph2Seq: Lane Topology Extraction with Language Model via Vertex-Edge Encoding and Connectivity Enhancement
This addresses the need for precise road structure understanding in autonomous driving, representing an incremental improvement over existing Transformer-based methods by better modeling edge information.
The paper tackled the problem of accurately extracting lane graphs for autonomous driving by introducing LaneGraph2Seq, which uses a language model with vertex-edge encoding and connectivity enhancement, achieving superior performance on nuScenes and Argoverse 2 datasets compared to state-of-the-art methods.
Understanding road structures is crucial for autonomous driving. Intricate road structures are often depicted using lane graphs, which include centerline curves and connections forming a Directed Acyclic Graph (DAG). Accurate extraction of lane graphs relies on precisely estimating vertex and edge information within the DAG. Recent research highlights Transformer-based language models' impressive sequence prediction abilities, making them effective for learning graph representations when graph data are encoded as sequences. However, existing studies focus mainly on modeling vertices explicitly, leaving edge information simply embedded in the network. Consequently, these approaches fall short in the task of lane graph extraction. To address this, we introduce LaneGraph2Seq, a novel approach for lane graph extraction. It leverages a language model with vertex-edge encoding and connectivity enhancement. Our serialization strategy includes a vertex-centric depth-first traversal and a concise edge-based partition sequence. Additionally, we use classifier-free guidance combined with nucleus sampling to improve lane connectivity. We validate our method on prominent datasets, nuScenes and Argoverse 2, showcasing consistent and compelling results. Our LaneGraph2Seq approach demonstrates superior performance compared to state-of-the-art techniques in lane graph extraction.