M
Mike Roberts
Researcher at Stanford University
Publications - 20
Citations - 1895
Mike Roberts is an academic researcher from Stanford University. The author has contributed to research in topics: Level set (data structures) & Segmentation. The author has an hindex of 15, co-authored 19 publications receiving 1509 citations. Previous affiliations of Mike Roberts include Harvard University & University of Calgary.
Papers
More filters
Journal ArticleDOI
Saturated Reconstruction of a Volume of Neocortex
Narayanan Kasthuri,Kenneth J. Hayworth,Daniel R. Berger,Daniel R. Berger,Richard Schalek,José Angel Conchello,Seymour Knowles-Barley,Dongil Lee,Amelio Vazquez-Reina,Verena Kaynig,Thouis R. Jones,Mike Roberts,Josh Morgan,Juan Carlos Tapia,H. Sebastian Seung,William Gray Roncal,Joshua T. Vogelstein,Randal Burns,Daniel L. Sussman,Carey E. Priebe,Hanspeter Pfister,Jeff W. Lichtman +21 more
TL;DR: In this paper, the authors describe automated technologies to probe the structure of neural tissue at nanometer resolution and use them to generate a saturated reconstruction of a sub-volume of mouse neocortex in which all cellular objects (axons, dendrites, and glia) and many subcellular components (synapses, synaptic vesicles, spines, spine apparati, postsynaptic densities, and mitochondria) are rendered and itemized in a database.
Journal ArticleDOI
Large-Scale Automatic Reconstruction of Neuronal Processes from Electron Microscopy Images
Verena Kaynig,Amelio Vazquez-Reina,Amelio Vazquez-Reina,Seymour Knowles-Barley,Mike Roberts,Thouis R. Jones,Narayanan Kasthuri,Eric L. Miller,Jeff W. Lichtman,Hanspeter Pfister +9 more
TL;DR: In this paper, a random forest classifier is combined with an anisotropic smoothing prior in a Conditional Random Field framework to generate multiple segmentation hypotheses per image, which are then combined into geometrically consistent 3D objects by segmentation fusion.
Posted Content
Hypersim: A Photorealistic Synthetic Dataset for Holistic Indoor Scene Understanding.
Mike Roberts,Nathan Paczan +1 more
TL;DR: This work introduces Hypersim, a photorealistic synthetic dataset for holistic indoor scene understanding, and finds that it is possible to generate the entire dataset from scratch, for roughly half the cost of training a popular open-source natural language processing model.
Proceedings ArticleDOI
Submodular Trajectory Optimization for Aerial 3D Scanning
Mike Roberts,Shital Shah,Debadeepta Dey,Anh Truong,Sudipta N. Sinha,Ashish Kapoor,Pat Hanrahan,Neel Joshi +7 more
TL;DR: In this article, the authors present an automatic method to generate drone trajectories, such that the imagery acquired during the flight will later produce a high-fidelity 3D model.
Journal ArticleDOI
An interactive tool for designing quadrotor camera shots
TL;DR: The tool makes it possible for novices and experts to design compelling and challenging shots, and capture them fully autonomously, and is evaluated in a user study with novice and expert cinematographers.