Image Processing

Team Members

Permanent Members Non permanent Members

Research Themes

Theme 1: textures and random fields

  • Theoretical study and exact simulation of point processes of blue noise type
  • Non Gaussian procedural texture synthesis
  • Spatial layout synthesis of vectorial objects by simulation of parametric Gibbs processes
  • Procedural Gabor noise synthesis by example
  • Crossings of shot noise processes
  • Online serious game for the study of visual perception
  • Compact representation of Gaussian textures and random phase textures
  • Structural descriptors for the exploration of medical images
  • Mathematical models for structured textures
  • Online demonstrations of standard algorithms for texture synthesis
  • Spot detection on texture backgrounds
  • Turning-band methods for the simulation of anisotropic fractional Brownian fields
  • Parametric estimation of anisotropic fractional Brownian fields and of Gaussian fields with matrix auto similarity
  • Representation of random phase textures  and applications to the prevention of osteoporosis

Theme 2: Registration, tracking, motion analysis and change detection, medical imaging

  • Diffeomorphic models and models of currents for shape comparison
  • Diffeomorphic registration of surfaces and curves
  • Dimensionality reduction and large deformations for the study of the anatomical variability of the brain
  • Vector-valued reproducing kernels for diffeomorphic models and currents
  • Optimal transport and diffeomorphic models
  • Cranio-facial reconstruction
  • Classifying registration
  • Variability analysis
  • Electron paramagnetic resonance imaging
  • Change detection and classification in satellite imaging
  • Rigid motion detection
  • Trajectory detection
  • Segmentation and tracking in fluorescence microscopy
  • Real-time motion detection and depth estimation in videos
  • Estimation of the fundamental matrix in stereovision
  • 3D/video modeling
  • Gorgones  project (network extraction and study, tracking, 3D)
  • Medical applications:
    • Functional imaging of aorta
    • Quantification of iron overload
    • Myocardial perfusion

Theme 3: Image restoration and satellite imaging

  • Wavelets construction
  • Construction of directional wavelet bases
  • Periodic + smooth image decomposition
  • Phase coherence and image quality
  • Aliasing correction
  • Total variation image denoising and variants
  • Multiplicative noise
  • Electronic paramagnetic resonance imaging.
  • Image restoration
    • Restoration  of images sampled over irregular networks
    • Restoration of old movies
    • Restoration of medical images
    • Restoration of photographic images
    • Application of phase coherence to restoration
  • Satellite imaging
    • Simulator of satellite images without aliasing and application to stereoscopic vision
    • Study of band-width signals in interferometry
    • Measure of satellite microvibrations on a disparity map
    • Stereoscopic reconstruction
    • Reconstruction of SMOS images.

Theme 4: Structure detection in images

  • A contrario methods for image analysis
  • Pattern Theory
  • Histogram segmentation and color palette
  • Dectection of boundaries
  • Ringing and aliasing detection
  • Study of color distributions in natural images
  • Perceptually inspired variational methods for color image enhancement
  • Differential geometry applied to color spaces
  • High dynamic range imaging
  • Natural image statistics
  • Diagnosis and quantification in medical images
    • Functional imaging of aorta
    • Quantification of iron overload in myocardium
    • Myocardial perfusion
    • Thoracic imaging