工作內容:
• 負責基於視覺、毫米波、激光雷達等多感測器的目標識別、語義分割、目標追蹤等相關演算法的研發
• 負責模型訓練與演算法優化以及模型量化剪枝等工作,包含優化演算法準確度和運行速度
• 負責自動駕駛項目中感知演算法模組的部署,測試方案的制定和實施
• 負責與其他模組(如硬體、演算法、軟體等)的系統整合和調試
• 負責相關技術文件的撰寫
任職要求:
1. 具備目標檢測、語義分割、3D物體檢測等方向應用經驗;熟悉電腦視覺領域的常見演算法模型(如目標檢測、目標追蹤、圖像分割等演算法模型)的設計和優化,包括常用CNN,FPN、YOLO、SSD、SegNet、ENet等神經網路
2. 熟悉常用的深度學習開源框架,如TensorFlow、PyTorch、MxNet、Caffe2、Caffe等,至少對其中一種框架熟悉
3. 熟悉C/C++、Linux、CMake、Git等開發環境和工具
Job Description:
• Responsible for the research and development of target recognition, semantic segmentation, target tracking and other related algorithms based on multi-sensors such as Vision, Radar, LiDAR and so on.
• Responsible for model training, algorithm optimization, and model quantification pruning, including optimizing algorithm accuracy and running speed.
• Responsible for the deployment of perception algorithm modules, formulation and implementation of test plans in autonomous driving projects
• Responsible for system integration and debugging with other modules (hardware, algorithms, software, etc.)
• Responsible for the drafting of related technical documentation
Qualifications:
1. Experience in target detection, semantic segmentation, 3D object detection, etc.; Familiar with the design and optimization of common algorithm models in the field of computer vision (such as object detection, object tracking, image segmentation, etc. algorithm models), including commonly used CNN, FPN, YOLO, SSD, SegNet, ENet and other neural networks
2. Familiar with commonly used deep learning open source frameworks, such as TensorFlow, PyTorch, MxNet, Caffe2, Caffe, etc., and be familiar with at least one of them
3. Familiar with development environments and tools like C/C++, Linux, CMake, Git, etc.