Home Research Publications  Group   Teaching Code  Data   CV   Media 

  • Skip-thought Vectors

    Code & trained models to turn sentences into vectors (sent2vec).

  • SegDeepM

    Code, data & trained models for object class detection.

  • Clothing Parsing

    Code & features for clothing parsing.

  • Beat the MTurkers

    Code for pixel-level object labeling given 3D bounding boxes.

  • RGB-D Scenes

    Code & data for holistic indoor scene understanding in RGB-D.

  • Distributed S-SVM

    Parallel cutting plane S-SVM.

  • Holistic Scene Parsing

    Code for holistic scene understanding (joint object detection, scene-type classification, image labeling).

  • 3D Urban Scenes

    Datasets for 3D urban traffic scene understanding.

  • Tracking by Detection

    Hierarchical object tracker using DPM detector.

  • LOST

    Localization in OpenStreetMap data using only visual odometry.


    Lightweight MATLAB tool for semantic and instance labeling.


    MATLAB / C++ code for efficent large-scale stereo matching.


    MATLAB / C++ code for iterative closest point matching (2D/3D).


Prof. Raquel Urtasun

Department of Computer Science
University of Toronto
6 King's College Rd
Toronto, Ontario, M5S 3G4

Phone: +1 (416) 946-8482
Email: urtasun (at) cs (dot) toronto (dot) edu Fax: TBD


shareright © 2002 Phlash  

eXTReMe Tracker