TopoPoint: Enhance Topology Reasoning via Endpoint Detection in Autonomous Driving

• URL: http://arxiv.org/abs/2505.17771v1
• Date: Fri, 23 May 2025 11:42:54 GMT
• Title: TopoPoint: Enhance Topology Reasoning via Endpoint Detection in Autonomous Driving
• Authors: Yanping Fu, Xinyuan Liu, Tianyu Li, Yike Ma, Yucheng Zhang, Feng Dai,
• Abstract summary: TopoPoint is a novel framework that explicitly detects lane endpoints and jointly reasons over endpoints and lanes for robust topology reasoning.<n>During training, we independently initialize point and lane query, and proposed Point-Lane Self-Attention to enhance global context sharing.<n>During inference, we introduce Point-Lane Geometry Matching algorithm that computes distances between detected points and lanes to refine lane endpoints.
• Score: 10.889692793133385
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Topology reasoning, which unifies perception and structured reasoning, plays a vital role in understanding intersections for autonomous driving. However, its performance heavily relies on the accuracy of lane detection, particularly at connected lane endpoints. Existing methods often suffer from lane endpoints deviation, leading to incorrect topology construction. To address this issue, we propose TopoPoint, a novel framework that explicitly detects lane endpoints and jointly reasons over endpoints and lanes for robust topology reasoning. During training, we independently initialize point and lane query, and proposed Point-Lane Merge Self-Attention to enhance global context sharing through incorporating geometric distances between points and lanes as an attention mask . We further design Point-Lane Graph Convolutional Network to enable mutual feature aggregation between point and lane query. During inference, we introduce Point-Lane Geometry Matching algorithm that computes distances between detected points and lanes to refine lane endpoints, effectively mitigating endpoint deviation. Extensive experiments on the OpenLane-V2 benchmark demonstrate that TopoPoint achieves state-of-the-art performance in topology reasoning (48.8 on OLS). Additionally, we propose DET$_p$ to evaluate endpoint detection, under which our method significantly outperforms existing approaches (52.6 v.s. 45.2 on DET$_p$). The code is released at https://github.com/Franpin/TopoPoint.

Related papers

• Rethinking Lanes and Points in Complex Scenarios for Monocular 3D Lane Detection [21.886902999963727]
  We present a theoretical analysis alongside experimental validation to verify that current sparse lane representation methods contain inherent flaws.<n>To address this issue, we propose a novel patching strategy to completely represent the full lane structure.<n>To enhance the model's perception of lane structures, we propose the PointLane attention (PL-attention)
  arXiv  Detail & Related papers  (2025-03-08T14:45:49Z)
• TopoLogic: An Interpretable Pipeline for Lane Topology Reasoning on Driving Scenes [27.930213859199473]
  We propose an interpretable method for lane topology reasoning based on lane geometric distance and lane query similarity. Our approach significantly outperforms the existing state-of-the-art methods on the mainstream benchmark OpenLane-V2. Our proposed geometric distance topology reasoning method can be incorporated into well-trained models without re-training.
  arXiv  Detail & Related papers  (2024-05-23T16:15:17Z)
• LanePtrNet: Revisiting Lane Detection as Point Voting and Grouping on Curves [8.037214110171123]
  Lane detection plays a critical role in the field of autonomous driving. We propose a novel approach, LanePtrNet, which treats lane detection as a process of point voting and grouping on ordered sets. We conduct comprehensive experiments to validate the effectiveness of our proposed approach, demonstrating its superior performance.
  arXiv  Detail & Related papers  (2024-03-08T08:45:42Z)
• LaneSegNet: Map Learning with Lane Segment Perception for Autonomous Driving [60.55208681215818]
  We introduce LaneSegNet, the first end-to-end mapping network generating lane segments to obtain a complete representation of the road structure. Our algorithm features two key modifications. One is a lane attention module to capture pivotal region details within the long-range feature space. On the OpenLane-V2 dataset, LaneSegNet outperforms previous counterparts by a substantial gain across three tasks.
  arXiv  Detail & Related papers  (2023-12-26T16:22:10Z)
• Waypoint-Based Imitation Learning for Robotic Manipulation [56.4287610994102]
  Waypoint labeling is underspecified, and requires additional human supervision. We propose Automatic Waypoint Extraction (AWE) for imitation learning. AWE can increase the success rate of state-of-the-art algorithms by up to 25% in simulation and by 4-28% on real-world bimanual manipulation tasks.
  arXiv  Detail & Related papers  (2023-07-26T17:45:55Z)
• RCLane: Relay Chain Prediction for Lane Detection [76.62424079494285]
  We present a new method for lane detection based on relay chain prediction. Our strategy allows us to establish new state-of-the-art on four major benchmarks including TuSimple, CULane, CurveLanes and LLAMAS.
  arXiv  Detail & Related papers  (2022-07-19T16:48:39Z)
• A Keypoint-based Global Association Network for Lane Detection [47.93323407661912]
  Lane detection is a challenging task that requires predicting complex topology shapes of lane lines and distinguishing different types of lanes simultaneously. We propose a Global Association Network (GANet) to formulate the lane detection problem from a new perspective. Our method outperforms previous methods with F1 score of 79.63% on CULane and 97.71% on Tusimple dataset with high FPS.
  arXiv  Detail & Related papers  (2022-04-15T05:24:04Z)
• Object Localization under Single Coarse Point Supervision [107.46800858130658]
  We propose a POL method using coarse point annotations, relaxing the supervision signals from accurate key points to freely spotted points. CPR constructs point bags, selects semantic-correlated points, and produces semantic center points through multiple instance learning (MIL) In this way, CPR defines a weakly supervised evolution procedure, which ensures training high-performance object localizer under coarse point supervision.
  arXiv  Detail & Related papers  (2022-03-17T14:14:11Z)
• Key Points Estimation and Point Instance Segmentation Approach for Lane Detection [65.37887088194022]
  We propose a traffic line detection method called Point Instance Network (PINet) The PINet includes several stacked hourglass networks that are trained simultaneously. The PINet achieves competitive accuracy and false positive on the TuSimple and Culane datasets.
  arXiv  Detail & Related papers  (2020-02-16T15:51:30Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.