Job Responsibilities:

1. Develop stereo calibration algorithms (binocular/self-calibration) and structured light 3D imaging algorithms for dual-camera systems.
2. Optimize and deploy stereo vision AI models (stereo matching, depth estimation, visual segmentation) on embedded platforms (e.g., Jetson, FPGA).
3. Enhance real-time performance and accuracy of binocular imaging systems, addressing challenges like weak textures and occlusions.
4. Participate in camera hardware selection and imaging optimization, collaborating with optical/electrical teams for system integration.

Requirements:

1. Master’s degree or higher in Computer Science, Automation, Mathematics, or related fields.
2. 3+ years of practical experience (including competitions/internships) in stereo vision or structured light camera development.
3. Proficient in OpenCV, PCL, ROS, and programming languages (C++/Python); experience with CUDA/OpenCL optimization preferred.
4. Familiarity with:
   • Zhang’s calibration method and multi-sensor calibration.
   • Classical stereo matching algorithms (e.g., SGM, ELAS) and deep learning models (e.g., PSMNet, RAFT-Stereo).
5. Preferred:
   • Experience in model quantization/pruning for edge deployment.
   • Background in 3D reconstruction/SLAM projects.