Joaquin Zepeda - Researcher

Joaquin Zepeda

Researcher

Location:Rennes

Laboratory:

Research:

I am interested in various research topics including

  • Deep learning

  • Sparse representations

  • Image and video indexing

  • Image classification and search

  • Computer vision

  • Image processing

  • Machine learning

 

Short Bio: 

I obtained my degrees of Bachelor of Engineering (Electrical) and Master of Engineering from McGill University in Montreal, QC, Canada in 2003 and 2006, respectively. In my masters I developed a novel codec for bursty erasure correction.

I obtained my PhD in 2010 from INRIA in Rennes, France, on topics including sparse representations, dictionary learning, image compression, image description and content-based image search.

Since October 2010, I have been at Technicolor Research & Innovation in Rennes, France, where I'm currently a permanent researcher.

 

Publications:

  1. Xinrui Lyu, Joaquin Zepeda, and Patrick Pérez. Maximum Margin Linear Classiers in Unions of Subspaces. In British Machine Vision Conference (BMVC), 2016.
  2. Praveen Kulkarni, Frédéric Jurie, Joaquin Zepeda, Patrick Pérez, and Louis Chevallier. SPLeaP: Soft Pooling of Learned Parts for Image Classication. In European Conference on Computer Vision (ECCV), 2016.
  3. Himalaya Jain, Patrick Pérez, Remi Gribonval, Joaquin Zepeda, and Herve Jégou. Approximate search with quantized sparse representations. In European Conference on Computer Vision (ECCV), 2016.
  4. Cagdas Bilen, Joaquin Zepeda, and Patrick Pérez. The CNN News Footage Dataset: Enabling Supervision in Image Retrieval. In European Signal Processing Conference (EUSIPCO), 2016.
  5. Cagdas Bilen, Joaquin Zepeda, and Patrick Pérez. Supervised Learning Of Low-Rank Transforms For Image Retrieval. In IEEE International Conference on Image Processing (ICIP), 2016.
  6. Joaquin Zepeda and Pérez Patrick. Exemplar SVMs as Visual Feature Encoders. In Computer Vision and Pattern Recognition (CVPR), 2015.
  7. Joaquin Zepeda, Turkan Mehmet, and Dominique Thoreau. Block Prediction Using Approximate Template Matching (Oral). In European Signal Processing Conference (EUSIPCO), 2015.
  8. Praveen Kulkarni, Joaquin Zepeda, Frédéric Jurie, Patrick Perez, and Louis Chevallier. Max-Margin, Single-Layer Adaptation of Transferred Image Features. In BigVision Workshop, Computer Vision and Pattern Recognition (CVPR), 2015.
  9. Praveen Kulkarni, Joaquin Zepeda, Frederic Jurie, Patrick Pérez, and Louis Chevallier. Learning the Structure of Deep Architectures via l-1 Penalization. In British Machine Vision Conference (BMVC), 2015.
  10. Praveen Kulkarni, Joaquin Zepeda, Frederic Jurie, Patrick Perez, and Louis Chevallier. Hybrid Multi-Layer Deep CNN / Aggregator Feature for Image Classication. In IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2015.
  11. Xavier Burgos, Joaquin Zepeda, Francois Le Clerc, and Patrick Pérez. Pose and expression-coherent face recovery in the wild (Oral). In International Conference on Computer Vision (ICCV) Workshops, Robust Subspace Learning and Computer Vision, 2015.
  12. Cagdas Bilen, Joaquin Zepeda, and Patrick Pérez. Learning Sparsity Inducing Analysis Operators for Discriminative Similarity Metrics. In Signal Processing with Adaptive Sparse Structured Representations (SPARS), 2015.
  13. Aakanksha Rana, Joaquin Zepeda, and Patrick Perez. Feature Learning for the Image Retrieval Task. In Asian Computer Vision and Pattern Recognition (ACCV) Workshops, 2014.
  14. Praveen Kulkarni, Gaurav Sharma, Joaquin Zepeda, and Louis Chevallier. Transfer Learning via Attributes for Improved On-the-fly Classication. In IEEE Winter Conference on Applications of Computer Vision (WACV), 2014.
  15. André F. de Araujo, Fernando Silveira, Haricharan Lakshman, Joaquin Zepeda, Anmol Sheth, Patrick Pérez, and Bernd Girod. The Stanford / Technicolor / Fraunhofer HHI Video Semantic Indexing System. In TRECVID, 2013.
  16. Joaquin Zepeda, Christine Guillemot, and Ewa Kijak. Image compression using the Iteration-Tuned and Aligned Dictionary (Oral). In IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pages 793{796, 2011.
  17. Joaquin Zepeda, Christine Guillemot, and Ewa Kijak. Image Compression Using Sparse Representations and the Iteration-Tuned and Aligned Dictionary (Journal.). IEEE Journal of Selected Topics in Signal Processing, 5(5):1061{1073, 2011.
  18. Joaquin Zepeda, Christine Guillemot, and Ewa Kijak. The Iteration-Tuned Dictionary for Sparse Representations (Oral, 2nd best paper). In IEEE Workshop on Multimedia Signal Processing, 2010.
  19. Joaquin Zepeda, Christine Guillemot, and Ewa Kijak. Approximate nearest neighbors using sparse representations. In International Conference on Acoustics, Speech and Signal Processing, 2010.
  20. Joaquin Zepeda. Nouvelles méthodes de représentations parcimonieuses; Application à la compression et lindexation dimages. PhD thesis, Université de Rennes 1 (INRIA), 2010.
  21. Joaquin Zepeda, Ewa Kijak, and Christine Guillemot. SIFT-based local image description using sparse representations. In IEEE International Workshop on Multimedia Signal Processing, 2009.
  22. Joaquin Zepeda and Fabrice Labeau. Tandem Filter Bank-DFT Code for Bursty Erasure Correction. In IEEE Vehicular Technology Conference, 2006.
  23. Joaquin Zepeda. Tandem Filterbank / DFT Code for Bursty Erasure Correction. Master's thesis, McGill University, 2006.
  24. C. Morand, J. Benois-Pineau, J.-Ph. Domenger, J. Zepeda, E. Kijak, and C. Guillemot. Scalable Object-based Video Retrieval in HD Video Databases (Journal). Image Communication, Elsevier, 2006.
  25. S. Martins, J. Zepeda, B. Picard, P. Radziszewski, and D. Roy. Investigating On-The-Shell Acoustics. In Autogeneous and Semiautogeneous Grinding Technology Conference, 2006.

 

Filed Patents:

  • Method and apparatus for image classification with joint feature adaptation and classifier learning (US20160140425).
  • Method and apparatus for ranking 2D candidate images (EP20140306828).
  • Image descriptor for media content (EP2859505A1).
  • Annotation display assistance device and method of assisting annotation display (EP2950224A1).
  • US20160119628A1,A method and apparatus for encoding image features using a differentiable bag-of-words encoder (EP3012780A1).
  • Method and apparatus for image classification with joint feature adaptation and classifier learning (EP3029606A2).
  • Computer tool with sparse representation (US9244948).
  • Image descriptor for media content (WO2013182241A1).
  • Synchronized movie summary (WO2014001137A1).
  • Method of obtaining a mega-frame image fingerprints for image fingerprint based content identification, method of identifying a video sequence, and corresponding device (WO2014174058A1).
  • Method of classification of images and corresponding device (WO2015032670).
  • Method of determination of stable zones within an image stream, and portable device for implementing the method (WO2016008759A1).
  • Method and apparatus for image retrieval with feature learning (WO2016037844A1).
  • Image recognition using descriptor pruning (WO2016037848A1).
  • Face inpainting using piece-wise affine warping and sparse coding
  • WO2016075274 Methods, systems and apparatus for image recognition based on recursively determined exemplar-support vector machines (E-SVM) features (WO2016050729A1).
  • Accelerated support vector machine (SVM) learning using clustering (WO2016075293A1).
  • A family of electro-optical and opto-electrical transfer functions.
 

Latest publications

Article

Transfer learning via attributes based on-the-fly classification

Praveen Kulkarni, Gaurav Sharma, Joaquin Zepeda, Chevallier Louis

WACV winter conference 2014
February 14, 2014