Long-Term Tracking Through Failure Cases (bibtex)
by Karel Lebeda, Simon Hadfield, Jiri Matas, Richard Bowden
Abstract:
Long term tracking of an object, given only a single instance in an initial frame, remains an open problem. We propose a visual tracking algorithm, robust to many of the difficulties which often occur in real-world scenes. Correspondences of edge-based features are used, to overcome the reliance on the texture of the tracked object and improve invariance to lighting. Furthermore we address long-term stability, enabling the tracker to recover from drift and to provide redetection following object disappearance or occlusion. The two-module principle is similar to the successful state-of-the-art long-term TLD tracker, however our approach extends to cases of low-textured objects. Besides reporting our results on the VOT Challenge dataset, we perform two additional experiments. Firstly, results on short-term sequences show the performance of tracking challenging objects which represent failure cases for competing state-of-the-art approaches. Secondly, long sequences are tracked, including one of almost 30000 frames which to our knowledge is the longest tracking sequence reported to date. This tests the re-detection and drift resistance properties of the tracker. All the results are comparable to the state-of-the-art on sequences with textured objects and superior on non-textured objects. The new annotated sequences are made publicly available
Reference:
Long-Term Tracking Through Failure Cases (Karel Lebeda, Simon Hadfield, Jiri Matas, Richard Bowden), In Proceeedings, IEEE workshop on visual object tracking challenge at ICCV, IEEE, 2013. (Oral, Slides, Data)
Bibtex Entry:
@InProceedings{Lebeda13,
  Title                    = {Long-Term Tracking Through Failure Cases},
  Author                   = {Karel Lebeda and Simon Hadfield and Jiri Matas and Richard Bowden},
  Booktitle                = {Proceeedings, IEEE workshop on visual object tracking challenge at ICCV},
  Year                     = {2013},

  Address                  = {Sydney, Australia},
  Month                    = {2 -- 8 } # dec,
  Organization             = {IEEE},
  Pages                    = {153 -- 160},
  Publisher                = {IEEE},

  Abstract                 = {Long term tracking of an object, given only a single instance in an initial frame, remains an open problem. We propose a visual tracking algorithm, robust to many of the difficulties which often occur in real-world scenes. Correspondences of edge-based features are used, to overcome the reliance on the texture of the tracked object and improve invariance to lighting. Furthermore we address long-term stability, enabling the tracker to recover from drift and to provide redetection following object disappearance or occlusion. The two-module principle is similar to the successful state-of-the-art long-term TLD tracker, however our approach extends to cases of low-textured objects. Besides reporting our results on the VOT Challenge dataset, we perform two additional experiments. Firstly, results on short-term sequences show the performance of tracking challenging objects which represent failure cases for competing state-of-the-art approaches. Secondly, long sequences are tracked, including one of almost 30000 frames which to our knowledge is the longest tracking sequence reported to date. This tests the re-detection and drift resistance properties of the tracker. All the results are comparable to the state-of-the-art on sequences with textured objects and superior on non-textured objects. The new annotated sequences are made publicly available},
  Comment                  = {<font color="red">Oral</font>, <a href="http://cvssp.org/Personal/KarelLebeda/papers/VOT2013_pres.zip">Slides</a>, <a href="http://cvssp.org/data/YTLongTrack/">Data</a>},
  Crossref                 = {ICCV13},
  Doi                      = {10.1109/ICCVW.2013.26},
  Gsid                     = {12696396808737824400},
  Keywords                 = {edge detection;feature extraction;image sequences;image texture;object detection;object tracking;VOT Challenge dataset;drift resistance properties;edge-based features;failure cases;lighting;long term object tracking;nontextured objects;object disappearance;object redetection;occlusion;short-term sequences;tracked object texture;tracking sequences;two-module principle;visual tracking algorithm;Apertures;Image edge detection;Lighting;Robustness;Target tracking;Visualization;computer vision;edge;line correspondence;long-term tracking;low texture;visual tracking},
  Timestamp                = {2013.11.27},
  Url                      = {http://personal.ee.surrey.ac.uk/Personal/S.Hadfield/papers/Long-Term%20Tracking%20Through%20Failure%20Cases.pdf}
}
Powered by bibtexbrowser