Global ETD Search

Search theses and dissertations gathered from participating repositories worldwide. Every result links back to the library that holds it. No account is needed.

Results

Showing 1 to 20 of 31 for “"nuScenes"”.

  1. SYNERGISTIC APPROACH FOR 3D OBJECT DETECTION AND TRACKING: RADAR AND CAMERA INTEGRATION IN AUTONOMOUS DRIVING

    … increase for car detection up to 29.0% on the nuScenes validation dataset. ARC-BEV, a 3D radar-camera object detection algorithm with a spatial attention mechanism, integrates modalities in Bird’s Eye View (BEV) space, achieving 43.7% mAP on nuScenes benchmark. Advancements in 3D-MOT are …

    nus Repository record for SYNERGISTIC APPROACH FOR 3D OBJECT DETECTION AND TRACKING: RADAR AND CAMERA INTEGRATION IN AUTONOMOUS DRIVING (opens in a new tab) Download file for SYNERGISTIC APPROACH FOR 3D OBJECT DETECTION AND TRACKING: RADAR AND CAMERA INTEGRATION IN AUTONOMOUS DRIVING (opens in a new tab)

  2. Enhanced 3D Object Detection and Tracking in Autonomous Vehicles: An Efficient Multi-Modal Deep Fusion Approach

    … datasets such as Argoverse, Microsoft COCO, and nuScenes thoroughly evaluate the capabilities of NeXtDet and NeXtFusion. The results show that NeXtFusion excels in detecting small and distant objects compared to existing methods. Notably, NeXtFusion achieves a state-of-the-art mAP score of 0.473 …

    iupui Repository record for Enhanced 3D Object Detection and Tracking in Autonomous Vehicles: An Efficient Multi-Modal Deep Fusion Approach (opens in a new tab)

  3. Klasifikace vozidel s použitím radaru

    … z pěti radarů Continental ARS 408 v datasetu nuScenes jsou transformována do společného souřadnicového systému, seskupena s tolerancí 50 ms a propojena s anotacemi pomocí interpolace ohraničujících kvádrů z 2 Hz na 13 Hz. Z radarových shluků je extrahován vektor 19 příznaků, který je …

    brno-tech Repository record for Klasifikace vozidel s použitím radaru (opens in a new tab)

  4. Multi-modal 3D semantic sccupancy representation for autonomous driving

    … <p>The framework is validated on Occ3D-nuScenes, an extension of the nuScenes dataset with dense 3D occupancy annotations. Extensive experiments show that the proposed methods achieve competitive performance in controlled urban scenarios. The evaluations demonstrate improvements in …

    oxford Repository record for Multi-modal 3D semantic sccupancy representation for autonomous driving (opens in a new tab)

  5. RRPN Data-Centric: Potencializando a fusão radar-câmera com supervisão 2D de alta fidelidade

    … detecção em duas eta- pas treinado com o dataset nuScenes, uma das principais bases públicas para pesquisa em direção autônoma. Nesse conjunto de dados, as anotações bidimensionais são obtidas pela projeção de caixas delimitadoras tridimensionais sobre o plano da imagem, processo que pode …

    brazil-ufpe Repository record for RRPN Data-Centric: Potencializando a fusão radar-câmera com supervisão 2D de alta fidelidade (opens in a new tab)

  6. Spatial Distribution Data Augmentations for Long Range LiDAR Object Detection

    … these data augmentations have on the popular nuScenes dataset. All experiments are conducted on the two common benchmarking models PointPillars and CenterPoint. The work shows that GTDS can negatively impact the detection accuracy of less frequent classes in nuScenes, even if the overall …

    chalmers Repository record for Spatial Distribution Data Augmentations for Long Range LiDAR Object Detection (opens in a new tab)

  7. INTEGRATING LOGICAL RULES INTO DEEP LEARNING FOR TRAJECTORY PREDICTION

    … training. Experiments using LAformer model on nuScenes dataset show our framework reduces rule violation rates by 59.3 - 86.7% and improves Miss Rate by 3.6%. Our work demonstrates that integrating logical rules leads to predictions that are more rule-compliant and accurate.

    nus Repository record for INTEGRATING LOGICAL RULES INTO DEEP LEARNING FOR TRAJECTORY PREDICTION (opens in a new tab) Download file for INTEGRATING LOGICAL RULES INTO DEEP LEARNING FOR TRAJECTORY PREDICTION (opens in a new tab)

  8. Detecting road boundaries and drivable regions in challenging weather conditions

    … use it in the segmentation task. We augment the nuScenes dataset with fog and rain and use it for our validation. We achieve 20% and 18% better road boundary and drivable region detection in inclement weather.

    uoit Repository record for Detecting road boundaries and drivable regions in challenging weather conditions (opens in a new tab)

  9. Sparse2SOAP: Domain Adaptation for LiDAR-Based 3D Object Detection

    … objects. We evaluate the proposed method using nuScenes and Waymo datasets for Unsupervised Domain Adaptation (UDA) tasks. We observe an increase in mAP and AP for classes with many QS objects. To the best of our knowledge, we are the first to perform feature alignment between sparse and dense …

    waterloo Repository record for Sparse2SOAP: Domain Adaptation for LiDAR-Based 3D Object Detection (opens in a new tab)

  10. Fusion Based Object Detection for Autonomous Driving Systems

    … accuracy on challenging datasets like KITTI and nuScenes. Finally, to address the sparsity inherent in LiDAR data, we introduce a novel method that generates virtual LiDAR points from camera images, augmented with semantic labels to detect sparsely distributed and occluded objects effectively and …

    unt Repository record for Fusion Based Object Detection for Autonomous Driving Systems (opens in a new tab)

  11. Learned perception systems for self-driving vehicles

    … tracking data. DEFT raises the bar on the nuScenes monocular 3D tracking challenge, more than doubling the performance of the previous top method (3.8x on AMOTA, 2.1x on MOTAR). We analyze the difference in performance between DEFT and the next best-published method on nuScenes and find …

    colostate Repository record for Learned perception systems for self-driving vehicles (opens in a new tab)

  12. Equivariant Sensor Fusion for Autonomous Driving Object Detection

    … and geometric data augmentation on the nuScenes benchmark. The results show that full equivariance alone does not work well. Across both augmentation and no-augmentation settings, EquiFusion performs consistently worse than the BEVFusion baseline, indicating that the fully equivariant …

    mcmaster Repository record for Equivariant Sensor Fusion for Autonomous Driving Object Detection (opens in a new tab)

  13. SAMFusion3D: Self-adaptive multi-modality fusion for 3D object detection in autonomous driving

    … the embedding. Our experiments conducted on the nuScenes dataset prove that our model can perform with reduced computational complexity while maintaining results comparable to those of the baseline model.

    glasgow Repository record for SAMFusion3D: Self-adaptive multi-modality fusion for 3D object detection in autonomous driving (opens in a new tab) Download PDF for SAMFusion3D: Self-adaptive multi-modality fusion for 3D object detection in autonomous driving (opens in a new tab)

  14. Communication-aware Trajectory Prediction and Tracking for Connected and Autonomous Vehicles

    … of predictive states. Evaluated on the nuScenes and Argoverse 2 benchmarks and tested under realistic communication and observation loss, the proposed models achieve competitive multimodal forecasting accuracy. The attention-enhanced CAA-GMM further improves interaction modeling while …

    ucf Repository record for Communication-aware Trajectory Prediction and Tracking for Connected and Autonomous Vehicles (opens in a new tab)

  15. Sensor Fusion in Neural Networks For Object Detection

    … research work. The models were trained on the Nuscenes dataset, one of the biggest automotive datasets available today. Using the concepts of augmentation, hyperparameter optimization, sensor fusion, and annotation filters, the CRF-Net was trained to achieve an accuracy score that was 69.13% …

    iupui Repository record for Sensor Fusion in Neural Networks For Object Detection (opens in a new tab)

  16. Planning-Aware Trajectory Prediction: Routing Downstream Planning Risk into a Frozen-Perception Multimodal Predictor

    … a controlled, frozen-perception protocol on the nuScenes dataset spanning oracle and perceived input domains. The full gradient path separates from the control on all four conditions, with non-overlapping bootstrap 95% confidence intervals for a collision proxy at a displacement-controlled …

    mcmaster Repository record for Planning-Aware Trajectory Prediction: Routing Downstream Planning Risk into a Frozen-Perception Multimodal Predictor (opens in a new tab)

  17. Learning Human-Aware Strategies for Legible and Predictable Robot Navigation

    … was evaluated both in simulation using the nuScenes dataset and in real-world trials using the Clearpath Jackal robot. In the user study, participants interacted with the robot in predefined navigation scenarios, and their feedback was used to assess the perceived clarity and predictability …

    waterloo Repository record for Learning Human-Aware Strategies for Legible and Predictable Robot Navigation (opens in a new tab)

  18. TOWARDS AUTONOMOUS DRIVING IN DENSE, HETEROGENEOUS, AND UNSTRUCTURED TRAFFIC

    … such as the Waymo Open Motion, Argoverse, NuScenes datasets, as well as public leaderboards where our tracking approach resulted in achieving rank 1 among over a 100 methods. We apply human driver behavior modeling in planning and navigation at unsignaled intersections and highways …

    maryland Repository record for TOWARDS AUTONOMOUS DRIVING IN DENSE, HETEROGENEOUS, AND UNSTRUCTURED TRAFFIC (opens in a new tab)

  19. Semantic and Geometry-Aware 3D Reconstruction of Outdoor Scenes Using Low-Cost Vision Sensors

    … Experimental results on the Waymo Open and nuScenes datasets demonstrate that the proposed approach improves both geometric fidelity and rendering quality, yielding more accurate camera pose estimates (1-10% improvement in translational RMSE), and measurable gains in rendering metrics (4% …

    calgary Repository record for Semantic and Geometry-Aware 3D Reconstruction of Outdoor Scenes Using Low-Cost Vision Sensors (opens in a new tab)

  20. Multimodal spatio-temporal deep learning framework for 3D object detection in instrumented vehicles

    … using public datasets such as KITTI and nuScenes with Frustum-ConvNet as a baseline architecture. Secondly, an attention mechanism was employed along with convolutional-LSTM to extract spatial-temporal information from sequence data to improve 3D estimations and to aid the architecture in …

    dcu Download PDF for Multimodal spatio-temporal deep learning framework for 3D object detection in instrumented vehicles (opens in a new tab)

Page 1 of 2