Principal Engineer
Current►3D Depth Sensing- Created Structured Light Pattern Design & Performance Simulation- Developed fast stereo vision algorithm- Made Stereo calibration tool making at mass produced level- Implemented real time depth calculation using multi-core CPU, GPU- Ported and optimized depth algorithm for low cost embedded target(Intel myriad2, nVidia TX2, PX2)- 3D Reconstruction for static scene using single depth camera(ICP, Voxel, Ray marching on GPU)- 3D Reconstruction for dynamic scene using multi depth camera(Multi depth data calibration, Data fusion)- Ported and optimized 3D Reconstruction algorithm (Using OpenCL for GPU)- 3D sensor performance measurement(Lidar, Stereo Vision, Structured light and 2D TOF)- Real-time 3D stereo depth enhance using guided filtering►Machine Vision- Made Lens distortion compensation- Developed optical flow & clustering algorithm and to implement TI DSP(VAYU superset) for real-time processing- Invented and designed virtual input device using laser and illuminator & control, detection software- Face & Gaze detection ► Image Processing & H/W making- Made 3D Holographic storage: PCI-x card design, SLM(Spatial Light Modulator)Encoding, Camera Decoding S/W► DNN related- Tensorflow, Tensorflow-lite, Mediapipe, CUDNN, Keras, and so on.