End-to-end Lane Shape Prediction with Transformers
This addresses inefficiencies and context limitations in lane detection for autonomous driving systems, though it is incremental as it builds on existing transformer and lane modeling approaches.
The authors tackled lane detection for autonomous vehicles by proposing an end-to-end transformer-based method that directly outputs lane shape parameters, achieving state-of-the-art accuracy on the TuSimple benchmark with lightweight model size and fast speed.
Lane detection, the process of identifying lane markings as approximated curves, is widely used for lane departure warning and adaptive cruise control in autonomous vehicles. The popular pipeline that solves it in two steps -- feature extraction plus post-processing, while useful, is too inefficient and flawed in learning the global context and lanes' long and thin structures. To tackle these issues, we propose an end-to-end method that directly outputs parameters of a lane shape model, using a network built with a transformer to learn richer structures and context. The lane shape model is formulated based on road structures and camera pose, providing physical interpretation for parameters of network output. The transformer models non-local interactions with a self-attention mechanism to capture slender structures and global context. The proposed method is validated on the TuSimple benchmark and shows state-of-the-art accuracy with the most lightweight model size and fastest speed. Additionally, our method shows excellent adaptability to a challenging self-collected lane detection dataset, showing its powerful deployment potential in real applications. Codes are available at https://github.com/liuruijin17/LSTR.