Documents
Supplementary material
3DLaneFormer: Rethinking Learning Views for 3D Lane Detection
![](/sites/all/themes/dataport/images/light-567757_1920.jpg)
- DOI:
- 10.60864/d05g-ax08
- Citation Author(s):
- Submitted by:
- Kun Dong
- Last updated:
- 14 June 2024 - 10:11am
- Document Type:
- Supplementary material
- Categories:
- Keywords:
- Log in to post comments
Accurate 3D lane detection from monocular images is crucial for autonomous driving. Recent advances leverage either front-view (FV) or bird’s-eye-view (BEV) features for prediction, inevitably limiting their ability to perceive driving environments precisely and resulting in suboptimal performance. To overcome the limitations of using features from a single view, we design a novel dual-view cross-attention mechanism, which leverages features from FV and BEV simultaneously. Based on this mechanism, we propose 3DLaneFormer, a powerful framework for 3D lane detection.
It outperforms the latest BEV-based or FV-based approaches through extensive experiments on challenging benchmarks and thus verifies the necessity and benefits of utilizing features in both views.