职位来源于实习僧。
[6 months, 4-5 days/week] Name of Project: Pangu Objectives: - The intern will work as a LiDAR perception engineer in our internal data driven lidar perception pipeline and software development. The intern will be guided by team leader to do start-of-the-art algorithm research with our internal, large scale dataset. Main Tasks: - Research and develop cutting-edge BEV perception algorithms using camera and multi-sensor (LiDAR, radar, Camera) data for tasks like 3D object detection and tracking - Participate in the development and optimization of the auto-labeling pipeline to improve annotation accuracy and efficiency - Train, fine-tune, and evaluate deep learning models for multi-sensor fusion-based automatic label generation - Process and analyze large-scale multi-sensor datasets to identify patterns and improve model generalization - Develop and implement tools for visualizing, validating, and editing auto-generated labels Key Qualifications: - Understanding on LiDAR/Camera sensor principle. Experience of point cloud processing - Familiar with Python programming - Familiar with PyTorch, experience of building a data preprocess, training and model exporting python packages - Experience of Deep Learning based LiDAR object detection or segmentation algorithms - Familiar with point cloud processing algorithm such as filtering, clustering, segmentation and feature extraction - Hands-on experience with BEV perception algorithms and frameworks (e.g., BEVFormer, LSS) - Research or project experience in multi-sensor fusion (camera + LiDAR/radar) - Familiarity with auto-labeling, self-supervised learning, or related technologies - Experience with common CV libraries (OpenCV) and data processing tools - Good research background with publications in top conferences (CVPR, ICCV, ECCV, IROS, etc.) - Development or research experience on CUDA and TensorRT programming
职位薪资:
250-300/天
想要了解更多,可以访问公司官网:https://www.mercedes-