The Visual Learning Group researches methods to acquire models of the real-world from visual data, such as images, videos, and motion sequences. Most of this research lies at the boundary between computer vision, machine learning and computer animation. Projects include learning image representations for visual recognition - the problem of determining what is where in a picture - but also extracting models of human movement from video and motion capture data.


  1. new article to appear in IEEE TPAMI:
    Classemes and Other Classifier-based Features for Efficient Object Categorization,
    A. Bergamo, and L. Torresani.

  2. new article to appear at WACV 2014:
    AutoCaption: Automatic Caption Generation for Personal Photos,
    K. Ramnath, S. Baker, L. Vanderwende, M. El-Saban, S.N. Sinha, A. Kannan,
    N. Hassan, M. Galley, Y. Yang, D. Ramanan, A. Bergamo, and L. Torresani.

  3. new article accepted at ICLR 2014:
    EXMOVES: Classifier-based Features for Scalable Action Recognition,
    D. Tran, and L. Torresani.

  1. L. Torresani wins a Neukom 2014 CompX grant.

  1. Thanks to Nvidia Corporation for supporting our research with a hardware donation.

  2. New article published in Pattern Recognition (March 2014):
    Learning Discriminative Localization from Weakly Labeled Data,
    M.H. Nguyen, L. Torresani, F. de la Torre, and Carsten Rother.

  1. New paper published in the Machine Learning Journal (Feb. 2014):
    Learning what is where from unlabeled images: joint localization and clustering of foreground objects,
    A. Chandrashekar, L. Torresani, and R. Granger.

  1. Our research on image-assisted Web search was featured in the Irish Times:

       ‘Secret sauce’ for web image searches.

  1. The Dartmouth ran a story about our SIGIR13 work:
       Torresani enhances search engine with image analysis.

  1. New paper at SIGIR 2013:

    What Can Pictures Tell Us About Web Pages? Improving Document Search using
    S. Rodriguez-Vaamonde, L. Torresani, and A. Fitzgibbon.

  1. New paper at CVPR 2013:

    Leveraging Structure from Motion to Learn Discriminative Codebooks for Scalable
    Landmark Classification
    A. Bergamo, S. Sinha, and L. Torresani.

  1. New paper at MobiSys 2013:

    CarSafe App: Alerting Drowsy and Distracted Drivers using Dual Cameras on

    C.W. You, N. Lane, F. Chen, R. Wang, Z. Chen, T. Bao, Y. Cheng, M. Lin,
    L. Torresani, and A. Campbell.

  1. New paper published in IEEE TPAMI, Feb. 2013:
       A Dual Decomposition Approach to Feature Correspondence,
    L. Torresani, V. Kolmogorov, and C. Rother.


We thank the following sources for supporting our research.

Logo design by Christine Claudino