Henri Rebecq – 25th March 2014

Plan 

What is image matting?



Description of the studied algorithm



Implementation details & Results

What is image matting? Accurate foreground/background estimation in images  Determining both full and partial pixel coverage (alpha-channel) 

What is image matting? 

Observed image = convex combination of foreground and background

7 unknowns  3 equations 

Highly underconstrained problem

Main application

+

=

What is image matting? 

How to reduce complexity?  User guidance

BG ?

FG

Trimap

What is image matting? 

How to reduce complexity?  Assumptions on image statistics ○ Smooth zones ○ Correlation between nearby pixels…

Color sampling method 

for each unknown pixel z  sample nearby known Fz and Bz colors  estimate matte using compositing equation:

Color sampling method 

How to choose correctly (Fz,Bz)?  Sample multiple candidate points (sample set)  Choose best pair among all possible pairs

Generating sampling set Split FG and BG regions into subregions

Input image

User defined trimap

Generating sampling set Split FG and BG regions into subregions Perform two-level color and spatial clustering (GMM)

Generating sampling set Split FG and BG regions into subregions Perform two-level color and spatial clustering (GMM) Choose centroids as candidate samples

Why is this idea good? 

Comprehensive sample set: each color distribution appears in the sample set



True samples are never missed

Computing alpha matte 

for each unknown pixel z  select corresponding subregion (the closer to the boundary the smaller region)  try all (F,B) pairs and choose the one which best

“explains” pixel z

(which maximizes a certain objective function among all possible pairs)  Compute alpha using compositing equation

Objective function (Fi,Bi) are not too far from Iz

(Fi,Bi) can explain color Iz well

favors well-separated color distributions

Color-based term Observed pixel = good convex combination of (Fi,Bi)

Spatial term Selected pair should be close to observed pixel

Overlapping term Cohen’s “d value”

Selected candidates should come from well-separated color distributions

inversely proportional to overlap between distributions

Pre/post processing Used to refine the alpha matte  Pre-processing: known regions are expanded for pixels close to the boundary (wrt color and spatial distance)  Post-processing: laplacian method to smooth alpha matte (regions with high confidence have higher smoothing weights) 

Results 

Everything implemented in C++ using OpenCV  Region partitioning  Two level hiarchical clustering

 Brute-force objective maximization  Alpha-matte computing  Pre-post processing

Choice of parameters 

Paper not comprehensive enough: only a shallow description of the algorithm  Number of subregions? Size of each subregion? Fixed number of subregions (4), width grows quadratically

 Number of color/spatial clusters? (“number of

peaks in the histogram”…)

For each subregion, fixed number of color/spatial clusters

 Multi-dimensional overlapping term? L2-norm of per-component Cohen’s d value

Thank you for your attention Demonstration if time permits

Improving IMAGE matting USING COMPREHENSIVE ... - GitHub

Mar 25, 2014 - ... full and partial pixel coverage (alpha-channel) ... Choose best pair among all possible pairs ... confidence have higher smoothing weights) ...

2MB Sizes 5 Downloads 127 Views

Recommend Documents

image compression using deep autoencoder - GitHub
Deep Autoencoder neural network trains on a large set of images to figure out similarities .... 2.1.3 Representing and generalizing nonlinear structure in data .

Using SqlRender - GitHub
6. 3.5 Case sensitivity in string comparisons . .... would have had no clue the two fields were strings, and would incorrectly leave the plus sign. Another clue that.

Using FeatureExtraction - GitHub
Oct 27, 2017 - Venous thrombosis. 3. 20-24. 2. Medical history: Neoplasms. 25-29. 4. Hematologic neoplasm. 1. 30-34. 6. Malignant lymphoma. 0. 35-39. 7. Malignant neoplasm of anorectum. 0. 40-44. 9. Malignant neoplastic disease. 6. 45-49. 11. Maligna

Specification on Image Data File Version - GitHub
5.4.10 ShootingRecord heap ... the JFIF file format[1], as described below), sample software shall be provided openly to player vendors. ... In the separate "Decisions Concerning Extension" section, we define how various companies.

Instructions for using FALCON - GitHub
Jul 11, 2014 - College of Life and Environmental Sciences, University of Exeter, ... used in FALCON is also available (see FALCON_Manuscript.pdf. ). ... couraged to read the accompanying technical document to learn ... GitHub is an online repository

Image processing using linear light values and other image ...
Nov 12, 2004 - US 7,158,668 B2. Jan. 2, 2007. (10) Patent N0.: (45) Date of Patent: (54). (75) ..... 2003, available at , 5.

Species Identification using MALDIquant - GitHub
Jun 8, 2015 - Contents. 1 Foreword. 3. 2 Other vignettes. 3. 3 Setup. 3. 4 Dataset. 4. 5 Analysis. 4 .... [1] "F10". We collect all spots with a sapply call (to loop over all spectra) and ..... similar way as the top 10 features in the example above.

Improving Dependency Parsers using Combinatory ...
[email protected], 1tdeoskar,[email protected] ... Dependency parsers can recover much of the .... sues, as the data for Hindi is small. We provided.

Improving convergence time of BPNN Image ...
for constructing image compression and decompression system. ... The main drawback of BPNN architecture for image compression is slow convergence.

Improving Web Image Search by Bag-Based Reranking
such as mi-SVM can be readily incorporated into our bag-based reranking .... very likely that multiple relevant images are clustered in a pos- ..... The equality holds as the objective function is concave in ..... 32-GB random access memory).Missing:

underwater image enhancement using guided trigonometric ... - Name
distortion corresponds to the varying degrees of attenuation encountered by light ... Then, He et al. [9,. 10] proposed the scene depth information-based dark.