Duy Lê Source Confirmed

OverviewAI-generated summary

Duy Lê's research focuses on the application of deep learning and neural networks for complex detection and segmentation tasks. His recent publications include work on accurate and real-time 3D pedestrian detection using an efficient attentive pillar network, and robust multi-sensor fusion in 3D object detection and BEV segmentation utilizing diffusion models. Lê has also contributed to the development of datasets for multi-person pose estimation and tracking, such as JRDB-Pose, and open-world panoptic segmentation and tracking in crowded environments with JRDB-PanoTrack. Further research interests include the development of deep convolutional neural networks for medical image analysis, specifically for detecting atypical femur fractures from radiographs, and benchmarking approaches for long-context capable models in natural language processing. His work involves collaborations with researchers at the University of Arkansas at Fayetteville, including Tran-Dac-Thinh Phan, Andrew Lockett, and Michael T. Kidd.

Metrics

• h-index: 5
• Publications: 20
• Citations: 96