NASA Student Internships in Computer Vision The Intelligent Robotics Group (IRG) at the NASA Ames Research Center has two immediate openings for student interns (Junior, Senior, or graduate student) to assist development of computer vision software. Experience with 2D/3D computer vision, C++, and UNIX (Linux) software development is needed. Flexible hours (ranging from part-time to full-time) and duration are available. Due to NASA clearance processing time, US citizenship is required. NASA Ames is located at Moffett Field, California in the heart of Silicon Valley. Ames is a leader in information technology research, with a focus on intelligent systems, supercomputing, and networking. More than 3,000 personnel are employed at Ames. In addition, approximately 300 graduate students, cooperative education students, post-doctoral fellows, and university faculty work at the Center. If you are interested in applying, please send the following via email: 1. A current CV or resume 2. A short letter describing your background and motivation to Terry Fong ---------------------------------------------------------------------------- 3D RECONSTRUCTION OF THE MARTIAN SURFACE FROM ORBITAL IMAGERY The amount of image data of the surface of Mars is increasing rapidly. Each new mission to Mars generates an order of magnitude more raw data than its predecessor. Scientists and mission planners increasingly rely on autonomous, software-based data analysis techniques that locate the most relevent pieces of data and derive useful high-level information; all with a minimum of human intervention. The NASA Ames Intelligent Robotics Group is currently developing one such software package, the NASA Ames Stereo Pipeline, for autonomously reconstructing 3D surfaces from stereo imagery. The Ames Stereo Pipeline has been used to create 3D models from ground imagery obtained by the Mars Pathfinder and Mars Exploration Rovers (MER) rovers as well as orbital imagery taken by the Mars Global Surveyor (MGS) satellite. These models provide an intuitive, familiar 3D environment in which mission scientists and engineers can make critically important decisions such as landing site selection and robotic arm instrument placement. The Intelligent Robotics Group has a full-time student intern position open for a Junior, Senior, or Graduate level college student with a strong interest in robotics and computer vision. The student will be responsible for adapting the Ames Stereo Pipeline for use during upcoming NASA missions to Mars, starting with the Mars Reconnaissance Orbiter (MRO) mission which will be entering Mars orbit during the summer of 2006. The student will have the opportunity to work closely with members of the instrument team that designed the wide-angle, high resolution context imager on MRO. The applicant will also have the opportunity to help develop new stereo processing capablities that advance the state of the art in surface reconstruction for current NASA missions (e.g. one possible project would be to develop a wide baseline stereo model generation capability for the Mars Exploration Rover (MER) mission). This project will entail significant software engineering and C++ programming, but the main focus will be on applied computer vision and surface reconstruction. The student will have ample opportunity to identify and solve challenging engineering problems geared toward improving the performance and expanding the capabilities of the Ames Stereo Pipeline. At the end of the internship, the student will be encouraged to summarize his/her work in a research paper and submit it to a conference for publication. ---------------------------------------------------------------------------- GPU ACCELERATED IMAGE PROCESSESING AND COMPUTER VISION NASA receives an order of magnitude more information from each new mission it sends to Mars. A considerable amount of this data is from passive camera-based sensor systems mounted on rovers or orbiting spacecraft. Consequently, computer vision algorithms are of particular relevence to NASA mission specialists who need to process the image data as quickly and efficiently as possible. Unfortunately, computer vision algorithms tend to be compute intensive, making real-time interaction with image data sets difficult. Furthermore, the use of computer vision as a source of real-time, high-resolution sensory information onboard rovers is typically problematic due to the computational overhead and time delays these algorithms incur. The NASA Ames Intelligent Robotics Group seeks to hire a Junior, Senior, or Graduate level student intern to help us accelerate our machine vision toolkit, the NASA Ames Vision Workbench, by using the general purpose processing capabilities of modern computer graphics hardware. The Vision Workbench can perform complicated tasks such as (1) reconstructing 3D surfaces using stereo vision techniques, (2) stitching and projecting images into high dynamic range (HDR) 3D panoramas, and (3) selecting scale and lighting invariant image features that can be used for automated image alignment and target tracking across multiple images. Our goal is to gain an order of magnitude speed improvement by harnessing the power and programmability of today's graphics hardware. To this end, the applicant will be resposible for rewriting critical portions of the Vision Workbench using OpenGL shader language or NVidia's CG shader language. The internship will entail significant development work using C++ and shader languages, and will require a reasonable level of familiarity with OpenGL, Linux development, and the hardware architecture of modern graphics cards. The applicant should also have a strong interest in computer vision and image processing, and must be proficient in the use of linear algebra and discrete time signal processing techniques. At the end of the internship, the student will be encouraged to summarize his/her work in a research paper and submit it to a conference for publication. ----------------------------------------------------------------------------