Prior Based Online Lane Graph Extraction from Single Onboard Camera Image
Not on the map. Why?
Abstract (arXiv)
The local road network information is essential for autonomous navigation. This information is commonly obtained from offline HD-Maps in terms of lane graphs. However, the local road network at a given moment can be drastically different than the one given in the offline maps; due to construction works, accidents etc. Moreover, the autonomous vehicle might be at a location not covered in the offline HD-Map. Thus, online estimation of the lane graph is crucial for widespread and reliable autonomous navigation. In this work, we tackle online Bird's-Eye-View lane graph extraction from a single onboard camera image. We propose to use prior information to increase quality of the estimations. The prior is extracted from the dataset through a transformer based Wasserstein Autoencoder. The autoencoder is then used to enhance the initial lane graph estimates. This is done through optimization of the latent space vector. The optimization encourages the lane graph estimation to be logical by discouraging it to diverge from the prior distribution. We test the method on two benchmark datasets, NuScenes and Argoverse. The results show that the proposed method significantly improves the performance compared to state-of-the-art methods.
Key facts. Cited by 2 papers in the map (data through 7 Oct 2026). Builds on 44 of its 59 references that are in the map. First posted on arXiv 25 Jul 2023.
Builds on
44 of 59 references.
References · 44
- 2019nuScenes: A multimodal dataset for autonomous drivingcited by 4,485
- 2015Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learningcited by 4,416
- 2019Analyzing and Improving the Image Quality of StyleGANcited by 3,580
- 2017Multi-Task Learning Using Uncertainty to Weigh Losses for Scene Geometry and Semanticscited by 1,345
- 2020Lift, Splat, Shoot: Encoding Images From Arbitrary Camera Rigs by Implicitly Unprojecting to 3Dcited by 1,040
- 2016Automatic chemical design using a data-driven continuous representation of moleculescited by 762
- 2019Argoverse: 3D Tracking and Forecasting with Rich Mapscited by 728
- 2018ADVENT: Adversarial Entropy Minimization for Domain Adaptation in Semantic Segmentationcited by 623
- 2018ChauffeurNet: Learning to Drive by Imitating the Best and Synthesizing the Worstcited by 391
- 2017Wasserstein Auto-Encoderscited by 371
- 2019Learning by Cheatingcited by 361
- 2018Multimodal Trajectory Predictions for Autonomous Driving using Deep Convolutional Networkscited by 289
32 more
- 2021DatasetGAN: Efficient Labeled Data Factory with Minimal Human Effortcited by 264
- 2022Cross-view Transformers for real-time Map-view Semantic Segmentationcited by 264
- 2015Bayesian SegNet: Model Uncertainty in Deep Convolutional Encoder-Decoder Architectures for Scene Understandingcited by 218
- 2016LSTM-based Encoder-Decoder for Multi-sensor Anomaly Detectioncited by 213
- 2021MP3: A Unified Model to Map, Perceive, Predict and Plancited by 211
- 2020Predicting Semantic Map Representations from Images using Pyramid Occupancy Networkscited by 175
- 2019Cross-view Semantic Segmentation for Sensing Surroundingscited by 173
- 2017Image to Image Translation for Domain Adaptationcited by 171
- 2018Towards End-to-End Lane Detection: an Instance Segmentation Approachcited by 129
- 2019Self-Ensembling with GAN-based Data Augmentation for Domain Adaptation in Semantic Segmentationcited by 126
- 2019Rules of the Road: Predicting Driving Behavior with a Convolutional Model of Semantic Interactionscited by 117
- 2020MotionNet: Joint Perception and Motion Prediction for Autonomous Driving Based on Bird's Eye View Mapscited by 100
- 2021Structured Bird's-Eye-View Traffic Scene Understanding from Onboard Imagescited by 98
- 2020Key Points Estimation and Point Instance Segmentation Approach for Lane Detectioncited by 77
- 2020FISHING Net: Future Inference of Semantic Heatmaps In Gridscited by 55
- 2021Improving black-box optimization in VAE latent space using decoder uncertaintycited by 51
- 2020DAGMapper: Learning to Map by Discovering Lane Topologycited by 49
- 2018Learning to Segment Medical Images with Scribble-Supervision Alonecited by 48
- 2020Hierarchical Recurrent Attention Networks for Structured Online Mapscited by 38
- 2019End-to-end Lane Detection through Differentiable Least-Squares Fittingcited by 36
- 2020Convolutional Recurrent Network for Road Boundary Extractioncited by 33
- 2016Anomaly Detection in Video Using Predictive Convolutional Long Short-Term Memory Networkscited by 24
- 2019Exploiting Sparse Semantic HD Maps for Self-Driving Vehicle Localizationcited by 24
- 2019Leveraging Crowdsourced GPS Data for Road Extraction from Aerial Imagerycited by 23
- 2018On the Latent Space of Wasserstein Auto-Encoderscited by 22
- 2021Decoder Fusion RNN: Context and Interaction Aware Decoders for Trajectory Predictioncited by 12
- 2018Iterative Deep Learning for Road Topology Extractioncited by 9
- 2020Learning High-Resolution Domain-Specific Representations with a GAN Generatorcited by 9
- 2020End-to-End Deep Structured Models for Drawing Crosswalkscited by 9
- 2020Action Sequence Predictions of Vehicles in Urban Environments using Map and Social Contextcited by 8
- 2021Statistical Regeneration Guarantees of the Wasserstein Autoencoder with Latent Space Consistencycited by 3
- 2020Dimensionality Reduction and Anomaly Detection for CPPS Data using Autoencodercited by 1
Built on by
Citing papers per year: 2023: 1; 2024: 1.
2 papers.