Description: Are you a computer vision expert with experience with deep learning? Matterport has the world’s largest dataset of aligned RGB-D images--200 million images from 220,000 different locations, and we want you to help us learn from it. We have a broad agenda for deep learning, including semantic labeling and segmentation, 3D object classification and pose estimation, depth from RGB, estimation of unseen 3D surfaces, texture/depth in-filling, keypoint matching, and dense stereo.
Matterport is a Series C startup that has developed a 3D camera and an end-to-end 3D platform that allows anyone to create, edit, and share 3D models of real-world spaces. Our tools are already used by thousands of professionals in a variety of industries, including real estate, construction, retail, hospitality, travel, and location-scouting, You can see some of the spaces captured by our customers at https://matterport.com/gallery/
Analyzing complex algorithms and developing new ones
Working with fuzzy, real-world data
Optimizing for performance when appropriate
We care the most about general ability to learn new things, but specific, directly relevant experience is, of course, desirable.
2+ years of experience in computer vision, 5+ preferred
Experience applying deep learning to computer vision problems
Familiarity with common deep neural net architectures
The ability to work independently and write clean, stable code
BS, MS, or PhD in Computer Science or a similar field
Familiarity with 3D graphics is a plus