Ferran Diego

Home page | Scientific Interests | Projects | Publications | Curriculum Vitae | Scholar Profile | Linked In Profile


Interdisciplinary Center for Scientific Computing (IWR)
Heidelberg Collaboratory for Image Processing (HCI)
Image Analysis and Learning Group
INF 205, 69120 Heidelberg

Tel.: +49 6221 5414831
e-mail: ferran.diego AT iwr.uni-heidelberg.de

Scientific Interests

My research interests are in the general areas of Computer Vision, Machine Learning, Image Analysis, Deep Learning, Multidimensional Image Analysis, Signal Processing, Big Data Analysis, Computer Science and Neuroscience.
Specifically, I am interested in the understanding, learning and probabilistic modeling of multi-dimensional data. Current projects focus on
  • Visual Scene Understanding based on Deep Learning
  • Probabilistic Graphical Models for Image Analysis
  • Structured Prediction
  • Convolutional Coding
  • Sparse Representations: Structured Sparsity and Group L1-regularization
  • Convex optimization
  • Learning and Inference for Aligning Data Sequences
  • Ensemble Methods


Structured Regression Gradient Boosting
It consists of a new way to train a structured output prediction model. More specifically, we train nonlocal data terms in a Gaussian Conditional Random Field (GCRF) by a generalized version of gradient boosting. The approach is evaluated on three challenging regression benchmarks: vessel detection, single image depth estimation and image inpainting. These experiments suggest that the proposed boosting framework matches or exceeds the state-of-the-art. [webpage]
Structured Learning for identifying co-activations of cells and neuronal assemblies
It consists of monitoring the spatio-temporal co-activation of neurons in rapidly oscillating three-dimensional networks based on protein-based fluorescent calcium images. This entails the detection and classification of cell centroids, and of calcium transients (events) that reappeared during different activity periods. we suggest addressing the whole problem by structured sparse dictionary learning since the events of interest are sparsely distributed in time and space.
Probabilistic Alignment of Video Sequences recorded by Moving Cameras
It consists of integrating multiple video sequences recorded independently into a single video sequence. This means to register both in time (synchronize frames) and space (image registration) so that the two videos sequences can be fused or compared pixel-wise. We focus mainly on analyzing and developing a probabilistic framework for video alignment, that is, a principled way to integrate multiple observations and prior information. [Thesis webpage | Video Ground Truthing | VA Framework | Subframe VA | Joint VA | Slice Matching | Road Segmentation | Geolocalization ]


See also my complete publication list or the Full list at MIP group

  • Structured Regression Gradient Boosting
    F. Diego, F. A. Hamprecht
    CVPR 2016, poster [webpage]
  • Sparse Space-Time Deconvolution for Calcium Image Analysis
    F. Diego, F. A. Hamprecht
    NIPS 2014, poster spotlight
  • Learning to disambiguate indistinguishable objects over time using weakly supervised structured learning
    L. Fiaschi, F. Diego, K. Gregor, M. Schiegg, U. Köthe, M. Zlatic, F. A. Hamprecht
    CVPR 2014, oral
  • Learning Multi-Level Sparse Representation
    F. Diego, F. A. Hamprecht
    NIPS 2013
  • Automated Identification of Neuronal Activity from Calcium Imaging by Sparse Dictionary Learning
    F. Diego, S. Reichinnek, M. Both, F. A. Hamprecht
    ISBI 2013
  • Road Geometry Classification by Adaptive Shape Models
    J. M. Alvarez, T. Gevers, F. Diego, A. M. Lopez
    IEEE Transactions on Intelligent Transportation Systems (ITS) [10.1109/TITS.2012.2221088]
  • Joint Spatio-Temporal Alignment of Sequences
    F. Diego, J. Serrat, A. M. Lopez
    IEEE Transactions on Multimedia (TMM) [10.1109/TMM.2013.2247390 | webpage]

Curriculum Vitae

Since 2012 PostDoc at Heidelberg Collaboratory for Image Processing (HCI) and CellNetworks
2008-2011 Granted PhD student (Training of University Professionals FPU Fellowship, Spanish Government in Computer Science ) at the Computer Vision Center, ADAS group, Universitat Autonoma de Barcelona
2007-2008 Teacher Assistant at the Computer Vision Center and Computer Science Dept., Universitat Autonoma de Barcelona
2006-2007 Granted M.S. Student (PIF scholarship) in Computer Science and Artifical Intelligence at the Computer Vision Center, ADAS group, Universitat Autonoma de Barcelona
2005-2006 Research Assistant at the Dept. Signal Theory and Communications, Speech Processing Group, Universitat Politecnica de Catalunya
2004-2005 M.S. Student in Speech Processing at the Dept. Signal Theory and Communications, Speech Processing Group, Universitat Politecnica de Catalunya
2000-2005 BD in Telecommunications at Universitat Politecnica de Catalunya