• Efficient Hierarchical Graph-Based Video Segmentation 

      Grundmann, Matthias; Kwatra, Vivek; Han, Mei; Essa, Irfan (Georgia Institute of TechnologyInstitute of Electrical and Electronics Engineers, 2010-06)
      We present an efficient and scalable technique for spatiotemporal segmentation of long video sequences using a hierarchical graph-based algorithm. We begin by oversegmenting a volumetric video graph into space-time ...
    • Geometric Context from Videos 

      Raza, S. Hussain; Grundmann, Matthias; Essa, Irfan A. (Georgia Institute of TechnologyInstitute of Electrical and Electronics Engineers, 2013-06)
      We present a novel algorithm for estimating the broad 3D geometric structure of outdoor video scenes. Leveraging spatio-temporal video segmentation, we decompose a dynamic scene captured by a video into geometric ...
    • Post-processing Approach for Radiometric Self-Calibration of Video 

      Grundmann, Matthias; McClanahan, Chris; Kang, Sing Bing; Essa, Irfan A. (Georgia Institute of TechnologyInstitute of Electrical and Electronics Engineers, 2013-04)
      We present a novel data-driven technique for radiometric self-calibration of video from an unknown camera. Our approach self-calibrates radiometric variations in video, and is applied as a post-process; there is no need ...
    • Video Segmentation by Tracking Many Figure-Ground Segments 

      Li, Fuxin; Kim, Taeyoung; Humayun, Ahmad; Tsai, David; Rehg, James M. (Georgia Institute of TechnologyInstitute of Electrical and Electronics Engineers, 2013-12)
      We propose an unsupervised video segmentation approach by simultaneously tracking multiple holistic figure-ground segments. Segment tracks are initialized from a pool of segment proposals generated from a figure-ground ...
    • Weakly Supervised Learning of Object Segmentations from Web-Scale Video 

      Hartmann, Glenn; Grundmann, Matthias; Hoffman, Judy; Tsai, David; Kwatra, Vivek; Madani, Omid; Vijayanarasimhan, Sudheendra; Essa, Irfan A.; Rehg, James M.; Sukthankar, Rahul (Georgia Institute of TechnologySpringer-Verlag Berlin / Heidelberg, 2012-10)
      We propose to learn pixel-level segmentations of objects from weakly labeled (tagged) internet videos. Specifically, given a large collection of raw YouTube content, along with potentially noisy tags, our goal is to ...