Optimizing and Learning for Super-resolution

Slides:



Advertisements
Similar presentations
Neural Networks and Kernel Methods
Advertisements

Recognising Panoramas M. Brown and D. Lowe, University of British Columbia.
Bayesian Belief Propagation
Active Appearance Models
Feature Based Image Mosaicing
Image Registration  Mapping of Evolution. Registration Goals Assume the correspondences are known Find such f() and g() such that the images are best.
DTAM: Dense Tracking and Mapping in Real-Time
Active Shape Models Suppose we have a statistical shape model –Trained from sets of examples How do we use it to interpret new images? Use an “Active Shape.
The fundamental matrix F
Improving resolution and depth of astronomical observations (via modern mathematical methods for image analysis) M. Castellano, D. Ottaviani, A. Fontana,
Summary of Friday A homography transforms one 3d plane to another 3d plane, under perspective projections. Those planes can be camera imaging planes or.
Investigation Into Optical Flow Problem in the Presence of Spatially-varying Motion Blur Mohammad Hossein Daraei June 2014 University.
Patch-based Image Deconvolution via Joint Modeling of Sparse Priors Chao Jia and Brian L. Evans The University of Texas at Austin 12 Sep
A Sampled Texture Prior for Image Super-Resolution Lyndsey C. Pickup, Stephen J. Roberts and Andrew Zisserman, Robotics Research Group, University of Oxford.
Reducing Drift in Parametric Motion Tracking
GrabCut Interactive Image (and Stereo) Segmentation Carsten Rother Vladimir Kolmogorov Andrew Blake Antonio Criminisi Geoffrey Cross [based on Siggraph.
Optimization & Learning for Registration of Moving Dynamic Textures Junzhou Huang 1, Xiaolei Huang 2, Dimitris Metaxas 1 Rutgers University 1, Lehigh University.
The loss function, the normal equation,
Classification and Prediction: Regression Via Gradient Descent Optimization Bamshad Mobasher DePaul University.
Computer Vision Optical Flow
Pattern Recognition and Machine Learning
Bayesian Image Super-resolution, Continued Lyndsey C. Pickup, David P. Capel, Stephen J. Roberts and Andrew Zisserman, Robotics Research Group, University.
Adam Rachmielowski 615 Project: Real-time monocular vision-based SLAM.
Motion Analysis (contd.) Slides are from RPI Registration Class.
CSci 6971: Image Registration Lecture 4: First Examples January 23, 2004 Prof. Chuck Stewart, RPI Dr. Luis Ibanez, Kitware Prof. Chuck Stewart, RPI Dr.
Recognising Panoramas
Direct Methods for Visual Scene Reconstruction Paper by Richard Szeliski & Sing Bing Kang Presented by Kristin Branson November 7, 2002.
ON THE IMPROVEMENT OF IMAGE REGISTRATION FOR HIGH ACCURACY SUPER-RESOLUTION Michalis Vrigkas, Christophoros Nikou, Lisimachos P. Kondi University of Ioannina.
Automatic Panoramic Image Stitching using Local Features Matthew Brown and David Lowe, University of British Columbia.
Lecture 1: Images and image filtering
Static Image Mosaicing
Optical flow and Tracking CISC 649/849 Spring 2009 University of Delaware.
Automatic Face Recognition for Film Character Retrieval in Feature-Length Films Ognjen Arandjelović Andrew Zisserman.
Automatic Image Alignment (feature-based) : Computational Photography Alexei Efros, CMU, Fall 2006 with a lot of slides stolen from Steve Seitz and.
CSCE 641 Computer Graphics: Image Registration Jinxiang Chai.
Rician Noise Removal in Diffusion Tensor MRI
A plane-plus-parallax algorithm Basic Model: When FOV is not very large and the camera motion has a small rotation, the 2D displacement (u,v) of an image.
Super-Resolution of Remotely-Sensed Images Using a Learning-Based Approach Isabelle Bégin and Frank P. Ferrie Abstract Super-resolution addresses the problem.
Efficient Model Selection for Support Vector Machines
By Yevgeny Yusepovsky & Diana Tsamalashvili the supervisor: Arie Nakhmani 08/07/2010 1Control and Robotics Labaratory.
Feature and object tracking algorithms for video tracking Student: Oren Shevach Instructor: Arie nakhmani.
1/20 Obtaining Shape from Scanning Electron Microscope Using Hopfield Neural Network Yuji Iwahori 1, Haruki Kawanaka 1, Shinji Fukui 2 and Kenji Funahashi.
CS 782 – Machine Learning Lecture 4 Linear Models for Classification  Probabilistic generative models  Probabilistic discriminative models.
Feature based deformable registration of neuroimages using interest point and feature selection Leonid Teverovskiy Center for Automated Learning and Discovery.
Esmaeil Faramarzi, Member, IEEE, Dinesh Rajan, Senior Member, IEEE, and Marc P. Christensen, Senior Member, IEEE Unified Blind Method for Multi-Image Super-Resolution.
1 University of Texas at Austin Machine Learning Group 图像与视频处理 计算机学院 Motion Detection and Estimation.
Raquel A. Romano 1 Scientific Computing Seminar May 12, 2004 Projective Geometry for Computer Vision Projective Geometry for Computer Vision Raquel A.
1 Markov random field: A brief introduction (2) Tzu-Cheng Jen Institute of Electronics, NCTU
EECS 274 Computer Vision Model Fitting. Fitting Choose a parametric object/some objects to represent a set of points Three main questions: –what object.
Bundle Adjustment A Modern Synthesis Bill Triggs, Philip McLauchlan, Richard Hartley and Andrew Fitzgibbon Presentation by Marios Xanthidis 5 th of No.
3.7 Adaptive filtering Joonas Vanninen Antonio Palomino Alarcos.
Lecture 9 Feature Extraction and Motion Estimation Slides by: Michael Black Clark F. Olson Jean Ponce.
Lucas-Kanade Image Alignment Iain Matthews. Paper Reading Simon Baker and Iain Matthews, Lucas-Kanade 20 years on: A Unifying Framework, Part 1
ICCV 2007 Optimization & Learning for Registration of Moving Dynamic Textures Junzhou Huang 1, Xiaolei Huang 2, Dimitris Metaxas 1 Rutgers University 1,
RECONSTRUCTION OF MULTI- SPECTRAL IMAGES USING MAP Gaurav.
Fast edge-directed single-image super-resolution
Summary of “Efficient Deep Learning for Stereo Matching”
Morphing and Shape Processing
10701 / Machine Learning.
Dynamical Statistical Shape Priors for Level Set Based Tracking
Probabilistic Models for Linear Regression
A guide to SR different approaches
PRAKASH CHOCKALINGAM, NALIN PRADEEP, AND STAN BIRCHFIELD
10701 / Machine Learning Today: - Cross validation,
Overfitting and Underfitting
Image and Video Processing
Support Vector Machines
The loss function, the normal equation,
Mathematical Foundations of BME Reza Shadmehr
Deblurring Shaken and Partially Saturated Images
Presentation transcript:

Optimizing and Learning for Super-resolution Lyndsey C. Pickup, Stephen J. Roberts & Andrew Zisserman Robotics Research Group, University of Oxford

The Super-resolution Problem Given a number of low-resolution images differing in: geometric transformations lighting (photometric) transformations camera blur (point-spread function) image quantization and noise. Estimate a high-resolution image:

Low-resolution image 1

Low-resolution image 2

Low-resolution image 3

Low-resolution image 4

Low-resolution image 5

Low-resolution image 6

Low-resolution image 7

Low-resolution image 8

Low-resolution image 9

Low-resolution image 10

Super-Resolution Image

Generative Model High-resolution image, x. y1 y2 y3 y4 Low-resolution images W4 W3 W2 W1 Registrations, lighting and blur.

Generative Model We don’t have: We have: Geometric registrations Point-spread function Photometric registrations Set of low-resolution input images, y.

Maximum a Posteriori (MAP) Solution y1 y2 y3 y4 W4 W3 W2 W1 x Standard method: Compute registrations from low-res images. Solve for SR image, x, using gradient descent. [Irani & Peleg ‘90, Capel ’01, Baker & Kanade ’02, Borman ‘04]

What’s new We solve for registrations and SR image jointly. We also find appropriate values for parameters in the prior term at the same time. Hardie et al. ’97: adjust registration while optimizing super-resolution estimate. Exhaustive search limits them to translation only. Simple smoothness prior softens image edges. i.e. given the low-res images, y, we solve for the SR image x and the mappings, W simultaneously. y1 y2 y3 y4 W4 W3 W2 W1 x

Overview of rest of talk Simultaneous Approach Model details Initialisation technique Optimization loop Learning values for the prior’s parameters Results on real images

Maximum a Posteriori (MAP) Solution y1 y2 y3 y4 W4 W3 W2 W1 x Image x. Corrupt with additive Gaussian noise. Warp, with parameters Φ. Blur by point-spread function. Decimate by zoom factor. y

Details of Huber Prior Huber function is quadratic in the middle, and linear in the tails. ρ (z,α) p (z|α,v) Red: large α Blue: small α Probability distribution is like a heavy-tailed Gaussian. This is applied to image gradients in the SR image estimate.

Details of Huber Prior Advantages: simple, edge-preserving, leads to convex form for MAP equations. Solutions as α and v vary: Ground Truth α=0.1 v=0.4 α=0.05 v=0.05 α=0.01 v=0.01 α=0.01 v=0.005 Edges are sharper Too much smoothing Too little smoothing

Advantages of Simultaneous Approach Learn from lessons of Bundle Adjustment: improve results by optimizing the scene estimate and the registration together. Registration can be guided by the super-resolution model, not by errors measured between warped, noisy low-resolution images. Use a non-Gaussian prior which helps to preserve edges in the super-resolution image. Remember, the classical approach is…. Fix ‘n’ solve.

Overview of Simultaneous Approach Start from a feature-based RANSAC-like registration between low-res frames. Select blur kernel, then use average image method to initialise registrations and SR image. Iterative loop: Update Prior Values Update SR estimate Update registration estimate

ML-sharpened estimate Initialisation Average image Use average image as an estimate of the super-resolution image (see paper). Minimize the error between the average image and the low-resolution image set. Use an early-stopped iterative ML estimate of the SR image to sharpen up this initial estimate. ML-sharpened estimate

Optimization Loop Update prior’s parameter values (next section) Update estimate of SR image Update estimate of registration and lighting values, which parameterize W Repeat till converged.

Joint MAP Results Registration Fixed Joint MAP Decreasing prior strength

Learning Prior Parameters α, ν Split the low-res images into two sets: Use first set to obtain an SR image. Find error on validation set.

Learning Prior Parameters α, ν Split the low-res images into two sets: Use first set to obtain an SR image. Find error on validation set. But what if one of the validation images is mis-registered?

Learning Prior Parameters α, ν Instead, we select pixels from across all images, choosing differently at each iteration. We evaluate an SR estimate using the unmarked pixels, then use the forward model to compare the estimate to the red pixels.

Learning Prior Parameters α, ν Instead, we select pixels from across all images, choosing differently at each iteration. We evaluate an SR estimate using the unmarked pixels, then use the forward model to compare the estimate to the red pixels.

Learning Prior Parameters α, ν To update the prior parameters: Re-select a cross-validation pixel set. Run the super-resolution image MAP solver for a small number of iterations, starting from the current SR estimate. Predict the low-resolution pixels of the validation set, and measure error. Use gradient descent to minimise the error with respect to the prior parameters.

Results: Eye Chart MAP version: fixing registrations then super-resolving Joint MAP version with adaptation of prior’s parameter values

Results: Groundhog Day

Results: Groundhog Day The blur estimate can still be altered to change the SR result. More ringing and artefacts appear in the regular MAP version. Blur radius = 1 Blur radius = 1.4 Blur radius = 1.8 Regular MAP Simultaneous

Lola Rennt

Real Data: Lola Rentt

Real Data: Lola Rentt

Real Data: Lola Rentt

Real Data: Lola Rentt

Conclusions Joint MAP solution: better results by optimizing SR image and registration parameters simultaneously. Learning prior values: preserve image edges without having to estimate image statistics in advance. DVDs: Automatically zoom in on regions with a registrations up to a projective transform and with an affine lighting model. Further work: marginalize over the registration – see NIPS 2006.