Computer vision algorithm and engineering expert- XC

Job Description

1. Develop and implement algorithms for perception and understanding of scene structured elements, including but not limited to lane detection, traffic light and sign detection, Road topology and online map generation and updating
2. R&D and implementation of 3D reconstruction, SLAM, multi-modal fusion positioning, multi-mode sensor calibration algorithms for autonomous driving, etc.
3. R&D , implement in-car perception algorithms, such as driver/passenger feature point detection, driver/passenger behavior understanding, gaze tracking, in-car object detection, etc.
4. Develop and implement the above algorithms in the vehicle Embedded platform transplantation and optimization, AI inference engine and operator optimization, GPUCPU optimization of CV algorithm, etc.
5. Develop high-precision map services, etc.
6. Develop light maps for autonomous driving / Picture-less sensing solutions, including BEV sensing of moving and static targets, sensing of the topological relationship between traffic lights and lane lines, etc.
7. Develop autonomous driving grid networks, AI-based multi-sensor integrated sensing solutions, etc. < /li>
8. Research visual or graphic models.

Qualifications

1. Master's degree or above from a 985211 institution, computer science related major (artificial intelligence, computer science, electronic information engineering, etc.), at least 3 years of relevant work experience. Good at tracking and learning cutting-edge AI technologies in academia or industry, with independent analysis and problem-solving abilities
2. In computer vision, pattern recognition, embedded systems, machine learning and deep learning (such as CNN) Experienced, with good sense of innovation and team spirit
3. Familiar with data structures and commonly used algorithms, with strong algorithm implementation capabilities, and at least mastering one or more common development methods such as C/C++, Python, Matlab, etc. Language
4. Have background and experience in target detection and recognition, semantic segmentation based on deep learning, be familiar with one of the mainstream deep learning frameworks, such as: Caffe, Tensorflow, PyTorch, etc., and have good understanding of neural networks Theory and project experience, or theory and project experience in 3D reconstruction, SLAM, or project experience in CV algorithm optimization, deep learning operator optimization
5. Familiar with GPGPU, CUDA, cuDNN, TensorRT, etc.
6. Familiar with vector mapping, crowdsourcing mapping, static fusion, etc.
7. Familiar with map engine development, map services, etc.
8. Strong learning ability and strong interest in algorithm development Interested, serious and responsible in work, able to face challenging problems with enthusiasm, good sense of innovation and team spirit
9. Have Applicants with relevant background in autonomous driving or actual project development experience will be given priority, or those with experience in international top artificial intelligence conferences/journal paper publishing/winners in well-known competitions will be given priority.