Main Responsibilities

• Design and develop state-of-art AI algorithms for embedded devices, including but not limited to: 1)Image/Video Enhancement. 2)Face Detection/Tracking/Landmark locations. 3)Object Detection. 4)Image/Video Segmentation. 5)Object/Scene Classification etc.
  

• Performance optimization for AI algorithms.
  

• Cooperate with hardware team customizing model size and computing consumption.
  

• Reading papers and evaluating state-of-the-art methods, investigating new ideas for mobile applications.
  
  

Qualifications

• MS/PhD in computer vision, pattern recognition and AI.
  

• Hands-on experience in developing CNN-based algorithms.
  

• Experience of C/C++ and Python.
  

• Familiarity with popular deep learning frameworks, TensorFlow, Pytorch or Caffe etc.
  

• Great communication skills, Fluent in English reading and writing.
  

• Strong Publication Record is a plus.
  

• Familiarity with conventional computer vision algorithm is a plus.
  

• Good knowledge on GPU, NPU, CPU architecture is a plus.