Meta AR/VR Job | Research Intern, Computer Vision for Egocentric Representation Learning (PhD)
Job（岗位）: Research Intern, Computer Vision for Egocentric Representation Learning (PhD)
Type（岗位类型）: Hardware +1 more
Citys（岗位城市）: Redmond, WA
Date（发布日期）: 2024-10-3
Summary（岗位介绍）
Reality Labs (RL) brings together a world-class, cross-disciplinary science and engineering teams with the shared goal of developing the next generation of AI for AR technologies.
The primary goal of this internship is on applying methods from computer vision to model adaptive processes in the human visual system. We are specifically looking for computer vision researchers who have or would like to learn how to apply CV models toward a better understanding of how human vision works. In this role, you will be embedded in a multidisciplinary team of scientists and engineers exploring innovative approaches to using data from egocentric cameras and motion detectors (IMU) and knowledge about human visual processes to model image encoding by the eye and applying those models to predict the quality of images presented on AR displays.
Our internships are twelve (12) to sixteen (16), or twenty-four (24) weeks long and we have various start dates throughout the year.
Qualifications（岗位要求）
Currently has or is in the process of pursuing a PhD in machine learning, computer vision, applied statistics, computational neuroscience, or a related field
Experience with research involving defining problems, exploring solutions, and analyzing and presenting results
Proficiency in python and machine learning libraries (numpy, scikit-learn, scipy, pandas, matplotlib, tensorflow, pytorch)
Experience in approaches to using self-supervised learning for computer vision
Experience in understanding of approaches to characterize the statistics of natural images and/or how to use properties of network layers to characterize high level scene/image properties
Must obtain work authorization in the country of employment at the time of hire, and maintain ongoing work authorization during employment
Description（岗位职责）
Identify, develop, implement, and evaluate methods for learning robust representations that model human visual performance from egocentric video data
Learn, evaluate, and use spatially calibrated camera and image processing pipeline to estimate physical scene parameters (e.g. luminance & chromaticity as a function of visual direction)
Make use of Meta’s large infrastructure to scale and speed up experimentation
Write modular research code that can be reused in other contexts.
Collaborate with researchers with expertise in image processing and computer vision
Additional Requirements（额外要求）
Experience with deep metric learning / neural net embedding methods
Basic understanding of color image processing pipelines
Basic understanding of modern adaptive camera and image processing pipelines
Intent to return to degree program after the completion of the internship/co-op.
Proven track record of achieving significant results as demonstrated by first-authored publications in high-impact journals or conferences (e. g., SIGGRAPH, SIGGRAPH Asia, Eurographics, CVPR, ICCV, ECCV, Journal of Vision, etc.).
Interpersonal skills: cross-group and cross-culture collaboration
