Paper in ECCV Workshop 2012: “Weakly Supervised Learning of Object Segmentations from Web-Scale Videos”

October 7th, 2012 Irfan Essa Posted in Activity Recognition, Awards, Google, Matthias Grundmann, Multimedia, PAMI/ICCV/CVPR/ECCV, Papers, Vivek Kwatra, WWW No Comments »

Weakly Supervised Learning of Object Segmentations from Web-Scale Videos

  • G. Hartmann, M. Grundmann, J. Hoffman, D. Tsai, V. Kwatra, O. Madani, S. Vijayanarasimhan, I. Essa, J. Rehg, and R. Sukthankar (2012), “Weakly Supervised Learning of Object Segmentations from Web-Scale Videos,” in Proceedings of ECCV 2012 Workshop on Web-scale Vision and Social Media, 2012. [PDF] [DOI] [BIBTEX]
    @InProceedings{    2012-Hartmann-WSLOSFWV,
      author  = {Glenn Hartmann and Matthias Grundmann and Judy
          Hoffman and David Tsai and Vivek Kwatra and Omid
          Madani and Sudheendra Vijayanarasimhan and Irfan
          Essa and James Rehg and Rahul Sukthankar},
      booktitle  = {Proceedings of ECCV 2012 Workshop on Web-scale
          Vision and Social Media},
      doi    = {10.1007/978-3-642-33863-2_20},
      pdf    = {},
      title    = {Weakly Supervised Learning of Object Segmentations
          from Web-Scale Videos},
      year    = {2012}


We propose to learn pixel-level segmentations of objects from weakly labeled (tagged) internet videos. Speci cally, given a large collection of raw YouTube content, along with potentially noisy tags, our goal is to automatically generate spatiotemporal masks for each object, such as dog”, without employing any pre-trained object detectors. We formulate this problem as learning weakly supervised classi ers for a set of independent spatio-temporal segments. The object seeds obtained using segment-level classi ers are further re ned using graphcuts to generate high-precision object masks. Our results, obtained by training on a dataset of 20,000 YouTube videos weakly tagged into 15 classes, demonstrate automatic extraction of pixel-level object masks. Evaluated against a ground-truthed subset of 50,000 frames with pixel-level annotations, we con rm that our proposed methods can learn good object masks just by watching YouTube.

Presented at: ECCV 2012 Workshop on Web-scale Vision and Social Media, 2012, October 7-12, 2012, in Florence, ITALY.



AddThis Social Bookmark Button

Essa, Egerstedt Named IEEE Fellows | School of Interactive Computing

November 21st, 2011 Irfan Essa Posted in Awards, In The News No Comments »

Via Georgia Tech School of Interactive Computing‘s Website > Essa, Egerstedt Named IEEE Fellows.

The IEEE Board of Directors has elected professors Irfan Essa and Magnus Egerstedt (both Interactive Computing) as Fellows in its Class of 2012.

Essa is a professor whose research focus is in computer vision, computer graphics, computational perception, robotics and computer animation. In his Fellow citation, Essa was lauded for “contributions to computer vision and graphics.”

“I feel honored to be selected to be part of a group of my peers that I respect and who have made amazing contributions to their fields,” Essa said. “I am glad that my contributions to computer vision and graphics are considered worthy for this honor, and I intend to continue working on my multi-disciplinary research.”

Egerstedt, an adjunct faculty member in the School of Interactive Computing with a primary appointment in the School of Electrical and Computer Engineering, works in optimal control, as well as modeling and analysis of hybrid and discrete event systems, with emphasis on motion planning and control of (teams of) mobile robots. His IEEE citation acknowledged “contributions to hybrid and networked control, with applications in robotics.”

Both professors are affiliated with the Robotics & Intelligent Machines (RIM) Center.

The IEEE Grade of Fellow is conferred by the Board of Directors upon those members with extraordinary records of accomplishment in any IEEE field of interest. IEEE Fellow is the highest grade of membership and is recognized by the technical community as a prestigious honor and an important career achievement. For a full list of the Fellow Class of 2012, visit the IEEE website.

AddThis Social Bookmark Button

PhD Fellowships from Google Research for Matthias Grundmann

May 16th, 2011 Irfan Essa Posted in Awards, In The News, Matthias Grundmann No Comments »

Congratulations to Matthias Grundmann, winner of the Google PhD Fellowship in Computer Vision for 2012.

via PhD Fellowships – Google Research.

Google PhD Fellowship Program Overview

Nurturing and maintaining strong relations with the academic community is a top priority at Google. The Google U.S./Canada PhD Student Fellowship Program was created to recognize outstanding graduate students doing exceptional work in computer science, related disciplines, or promising research areas. Last year we awarded 14 unique fellowships to some amazing students in the US and Canada:

  • Matthias Grundmann, Google U.S./Canada Fellowship in Computer Vision (Georgia Institute of Technology)
AddThis Social Bookmark Button