Uncertainty quantification methods and their applications in object detection / submitted by Marius Schubert, M.Sc. Wuppertal, October 10, 2023
Content
- Acknowledgments
- Foreword
- Contents
- Introduction
- Review of Basic Material
- Applications of Uncertainty Quantification Methods in Object Detection
- MetaDetect
- Active Learning Sandbox
- Identifying Label Errors
- Active Learning with Noisy Oracle
- LidarMetaDetect
- MetaDetect: Uncertainty Quantification and Prediction Quality Estimates for Object Detection
- Introduction
- The Generic Object Detection Pipeline
- Uncertainty Metrics for Object Detection
- Numerical Experiments: Pascal VOC, KITTI and COCO
- Conclusion and Outlook
- Towards Rapid Prototyping and Comparability in Active Learning for Deep Object Detection
- Introduction
- Related Work
- A Sandbox Environment with Datasets, Models and Evaluation Metrics
- Experiments
- Conclusion
- Identifying Label Errors in Object Detection Datasets by Loss Inspection
- Deep Active Learning with Noisy Oracle in Object Detection
- LMD: Light-weight Prediction Quality Estimation for Object Detection in Lidar Point Clouds
- Conclusion and Outlook
- Bibliography