Si Lu Email & Phone Number

Santa Clara, Ca, Us

Autonomous Vehicles: HD Map Generation• HD map fusion: design and implement lane boundary fusion.• HD map fusion: design and implement pole/sign/traffic lights fusion.• HD map emitting: implement the emitting of all fused map contents.• HD map rendering: construct the local layout for rendering and visualization.• HD map rendering: implement a graphics pipeline to render HD map contents.Autonomous Vehicles: Localization:• Localization: KPI tracking/visualizationand in Athena websites.• Localization: fast KPI computing in the cloud computing platform.• Localization: design/implement a high-accuracy camera-based localization• Localization: so far the first one to purly use low cost sensors (IMU, GPS ...).• Localization: design/implemente core algorithm for lidar/radar localization

Portland, Or, Us

(1) CS 447/547 - Computer Graphics (Instructor): gave lectures about movies, games, animations and 3d rendering. Helped students with two course projects - a mini-Photoshop (FLTK/OpenCV) with basic image processing and an Amusement Park 3D animation rendering project (OpenGL).(2) CS 510/610 - Computational Photography (Instructor): teach research topics ranging from concepts of digital camera and photography to computer vision/graphics techniques, including high dynamic range imaging, panorama stitching, image segmentation & matting, video stabilization, virtual reality basics, deep learning in computer vision etc.(3) CS 250/251 - Discrete Structures I & II: grade assignments, exams and answere student questions.(4) CS 581 - Theory of computation: grade assignments

Portland, Or, Us

Develop new computer vision and graphics techniques for digital images and videos. My research interests includes novel view synthesis, video stabilization, image stitching, Image denoising and depth enhancement.Projects: (1) Novel View Synthesis ( C/C++/Matlab/OpenFrameworks/Maya ) Developed an algorithm to generate novel views for multi-camera arrays. By using feature matching, image warping and image-based rendering techniques, plausible and parallax-free novel views can be robustly generated for scenes with large camera and object motions. (submitted to CVPR 2018)(2) Patch Matching for Image Denoising ( C/C++/Matlab/OpenCV/Python/Pytorch ): Developed a clustering-based approach to consistently improve patch-based denoising techniques' (like BM3D) performance by using learned patch descriptors trained via deep learning. (submitting to ECCV 2018)(3) Depth Enhancement ( C/Matlab/OpenCV/OpenNI/Kinect SDK/FLTK) Developed a depth enhancement algorithm that performs depth map completion and denoising simultaneously for RGBD cameras like Microsoft Kinect and achieves appealing depth filling and denoising results (CVPR 2014 paper). (4) Flash Light Detection for Unmanned Aerial Vehicles: Company supported projects Developed an algorithm that allows real-time flash light detection for Unmanned Aerial Vehicles. By designing specific color-based features, we achieved a 89.2% detection rate with a small false positive rate (0.6%).

Mountain View, California, Us

Multi-view 3D reconstruction ( C/C++/OpenCV ): Developed 3D reconstruction enhancement algorithms for Virtual Reality videos captured by light field camera arrays using patch matching, geodesic path expanding and sensor fusion. Integrated into Lytro production pipeline.

北京, Beijing, Cn

Robocup Humanoid League Competition: In the Humanoid League, autonomous robots with a human-like body plan and human-like senses play soccer against each other. Unlike humanoid robots outside the Humanoid League the task of perception and world modeling is not simplified by using non-human like range sensors. In addition to soccer competitions technical challenges take place. Dynamic walking, running, and kicking the ball while maintaining balance, visual perception of the ball, other players, and the field, self-localization, and team play are among the many research issues investigated in the Humanoid League.Projects:(1) Robot Vision System I: Fast-field line detecting II: Objects detecting III: Wireless vision parameter calibration interface IV: Ball locating and self-locating(2) Structure and Gaits I: Body Structure Design II: Gaits designingPlease go to " http://web.cecs.pdx.edu/~lusi/Robocup_files/Robocup.htm " for details.

Doctor Of Philosophy (Ph.D.), Computer Science

Master Of Science - Ms, Computer Science

Master Of Science (M.S.), Fluid Mechaincs

Bachelor'S Degree, Mechaincs