Marie Curie Project: Learning to Combine Hierarchical Image Modeling with 2-D Segmentation and 3-D Pose Recovery of Visual Objects
This project is funded by the European Commission; Marie Curie IIF, contract number PIIF-GA-2009-236173, acronym IMASEG3D
Researcher: Gustavo Carneiro. CI: Joao Paulo Costeira

  General Objectives

  1. Improve accuracy of image retrieval and annotation systems
  2. Efficient (fast) search methods with hierarchical models
  3. 3-D model can start providing full scene understanding - opens up new research problem

  Timeline

  Achievements

  1. 1 IEEE TIP paper, 1 IEEE TPAMI submission (under review), 2 journal papers in preparation
  2. 13 conference papers in top computer vision, image processing and machine learning conferences
  3. 3 versions of a 2D segmentation system
  4. 1 image feature learning system
  5. 1 image annotation and retrieval system
  6. Participation in 5 international conferences ([ISBI 2010], [CVPR 2010], [ICPR 2010], [ICMR 2011], [ICCV 2011],)
  7. Outstanding reviewer award by the committee of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR). San Francisco, USA. 2010 (top-ranked conference in computer vision).
  8. Invited for the [50,000th Marie Curie Fellowship Event].
  9. [Invited lecture - Back to Basics Colloquium at the University of Porto], Portugal. May, 2010.
  10. Active participation in the following two projects within the IST: a) [PRINTART] (Image retrieval and annotation of art images); and b) HEARTTRACK (Segmentation and tracking of the left ventricle of the heart in medical images and volumes)
  11. Supervision of post-doctoral fellow Nuno Miguel Pinho da Silva
  12. Supervision of MSc student Joao Tiago Fonseca
  13. Supervision of the researcher Duarte Lazaro in the PRINTART project
  14. Co-organized workshop 'Printart - Where Computer Vision meets Art'
  15. Worked in the editorial board of several journals, including IEEE TPAMI, TMI, TIP, TMM, etc.
  16. Worked in the program committee of several internation conferences, such as CVPR, ICCV, ICPR, etc.

  Journal publications

  1. Gustavo Carneiro, Jacinto Nascimento, Antonio Freitas. The Segmentation of the Left Ventricle of the Heart from Ultrasound Data using Deep Learning Architectures and Derivative-based Search Methods. IEEE Transactions on Image Processing. Vol. 21, No. 3. 2012.  [pdf]

  2. 1 IEEE TPAMI submission (under review), 2 journal papers in preparation.

  Conference publications

  1. Gustavo Carneiro, Jacinto C. Nascimento. Incremental On-line Semi-supervised Learning for Segmenting the Left Ventricle of the Heart from Ultrasound Data.International Conference on Computer Vision (ICCV). Barcelona, Spain. 2011.  [pdf]
  2. Jacinto C. Nascimento, Gustavo Carneiro. Reducing the Training Set Using Semi-Supervised Self-Training Algorithm for Segmenting the Left Ventricle in Ultrasound Images.International Conference on Image Processing (ICIP). Brussels, Belgium. 2011.  [pdf]
  3. Gustavo Carneiro. Graph-based Methods for the Automatic Annotation and Retrieval of Art Prints.ACM International Conference on Multimedia Retrieval (ICMR). Trento, Italy. 2011.  [pdf]
  4. Gustavo Carneiro, Jacinto C. Nascimento, Antonio Freitas. Semi-supervised self-training model for the segmentationof the left ventricle of the heart from ultrasound data.IEEE International Symposium on Biomedial Imaging (ISBI). Chicago, USA. 2011.  [pdf]
  5. Gustavo Carneiro, Joao Paulo Costeira. The automatic annotation and retrieval of digital images of prints and tile panels using network link analysis algorithms.SPIE Computer Vision and Image Analysis of Art II (Proceedings of SPIE Volume: 7869. San Francisco, USA. 2011.  [pdf]
  6. Nuno Pinho da Silva, Manuel Marques, Gustavo Carneiro, Joao Paulo Costeira. Explaining scene composition using kinematic chains of humans: application to Portuguese tiles history.SPIE Computer Vision and Image Analysis of Art II (Proceedings of SPIE Volume: 7869. San Francisco, USA. 2011.
  7. Ricardo S. Cabral, Joao Paulo Costeira, Fernando de La Torre, Alexandre Bernardino, Gustavo Carneiro. Time and order estimation of paintings based on visual features and expert priors.SPIE Computer Vision and Image Analysis of Art II (Proceedings of SPIE Volume: 7869. San Francisco, USA. 2011.
  8. Gustavo Carneiro. The Automatic Design of Feature Spaces for Local Image Descriptors using an Ensemble of Non-linear Feature Extractors. IEEE Conference on Computer Vision and Pattern Recognition (CVPR). San Francisco, USA. 2010.  [pdf]
  9. Gustavo Carneiro and Jacinto Nascimento. Multiple Dynamic Models for Tracking the Left Ventricle of the Heart from Ultrasound Data using Paricle Filters and Deep Learning Architectures. IEEE Conference on Computer Vision and Pattern Recognition (CVPR). San Francisco, USA. 2010.  [pdf]
  10. Gustavo Carneiro. A Comparison Study on the Use of an Ensemble of Feature Extractors for the Automatic Design of Local Image Descriptors. International Conference on Pattern Recognition (ICPR). Istanbul, Turkey. 2010. [pdf]
  11. Gustavo Carneiro and Jacinto Nascimento. The Fusion of Deep Learning Architectures and Particle Filtering Applied to Lip Tracking. International Conference on Pattern Recognition (ICPR). Istanbul, Turkey. 2010. [pdf]
  12. Gustavo Carneiro and Jacinto Nascimento. Robust Left Ventricle Segmentation from Ultrasound Data Using Deep Neural Networks and Efficient Search Methods.International Symposium on Biomedical Imaging (ISBI). Rotterdam, the Netherlands. 2010.  [pdf]
  13. Jacinto Nascimento and Gustavo Carneiro. Efficient Search Methods and Deep Belief Networks with Particle Filtering for Non-rigid Tracking: Application to Lip Tracking.International Conference on Image Processing (ICIP). Hong Kong. 2010.  [pdf]