plateform stringclasses 1
value | repo_name stringlengths 13 113 | name stringlengths 3 74 | ext stringclasses 1
value | path stringlengths 12 229 | size int64 23 843k | source_encoding stringclasses 9
values | md5 stringlengths 32 32 | text stringlengths 23 843k |
|---|---|---|---|---|---|---|---|---|
github | ornithos/lpsvm-master | frankWolfe.m | .m | lpsvm-master/Optim Ideas/frankWolfe.m | 4,364 | utf_8 | c20bcba19bb447d620c41e4095293ca5 | function [u, beta, exitflag] = frankWolfe(H, initial, D, maxIter, epsilon, ...
tol, searchType, dbg)
% Frank-Wolfe solution of the minimax problem for LPSVM
%
% Will output both the optimal u and the minimax achieved, beta.
%
% Arguments:
% H - (Matrix) rows = hypotheses / cols = datapoints
% ini... |
github | ornithos/lpsvm-master | SCDNNPrimal2old.m | .m | lpsvm-master/Optim Ideas/SCDNNPrimal2old.m | 2,329 | utf_8 | 0627c0e5d06f8aca4fe55e35c459ef0e | function x = SCDNNPrimal2a(f, A, b, lambda, mu, x0, maxIter, tol)
% (Stochastic) Coordinate Ascent for non-negative quadratic minimisation.
% (2) We now actually take analytic (optimal) step for each j rather than
% gradient step. Reduced compute time.
%
% Since we expect a sparse vector, we can reduce the probability ... |
github | ornithos/lpsvm-master | solveLPActiveAB2.m | .m | lpsvm-master/Optim Ideas/solveLPActiveAB2.m | 4,147 | utf_8 | 8aea67ef78cd372d8cc57049167605c1 | function[out, exitflag, primal, time, msgl] = ...
solveLPActiveAB2(H, D, bPrimal, tradeoff)
% SOLVELPACTIVEAB2 Solve LP approximately with greedy heuristic. Solving
% the LP exactly may not be necessary, since the new rows H_j created are
% only assumed to be weak learners, NOT the optimal rows to a... |
github | ornithos/lpsvm-master | frankWolfe3.m | .m | lpsvm-master/Optim Ideas/frankWolfe3.m | 4,030 | utf_8 | 126325dfcfa82d30c1cadca792a7e01e | function [u, beta, exitflag] = frankWolfe3(H, initial, D, maxIter, epsilon, ...
tol, searchType, dbg)
% Third Frank-Wolfe (Reverse) solution of the minimax problem for LPSVM.
% Here we have relaxed the problem to a continuous set consisting of all
% convex combinations of H, and used Von Neumann/Sion Theorem to vie... |
github | ornithos/lpsvm-master | augLagOpt2.m | .m | lpsvm-master/Optim Ideas/augLagOpt2.m | 1,616 | utf_8 | 6691e6c8e1106cdaffeec61079da5c52 | function [x, y] = augLagOpt2(f, A, b, x0, maxIter, tol, lambda)
% augLagOpt 2nd Augmented Lagrangian Optimisation of the LPSVM Dual Problem
% GULER
% This is similar to a Method of Multipliers approach
% Arguments:
% f - (Col Vector) Objective function
% A - (Matrix) equality constraint... |
github | ornithos/lpsvm-master | SCDNNQuad2a.m | .m | lpsvm-master/Optim Ideas/SCDNNQuad2a.m | 2,796 | utf_8 | c4adc351df6e4f4bf22fe4feaaaf24c6 | function x = SCDNNQuad2a(f, A, b, y, lambda, x0, maxIter, tol)
% (Stochastic) Coordinate Ascent for non-negative quadratic minimisation.
% (2) We now actually take analytic (optimal) step for each j rather than
% gradient step. Reduced compute time.
%
% Since we expect a sparse vector, we can reduce the probability of
... |
github | ornithos/lpsvm-master | augLagOpt.m | .m | lpsvm-master/Optim Ideas/augLagOpt.m | 1,722 | utf_8 | 3d0f6029a1bee64c7ea495c1e4158157 | function x = augLagOpt(f, A, b, x0, maxIter, tol, eps)
% augLagOpt Augmented Lagrangian Optimisation of the LPSVM Dual Problem
% EVTUSHENKO ET AL.
% This is similar to a Method of Multipliers approach
% Arguments:
% f - (Col Vector) Objective function
% A - (Matrix) equality constraint ... |
github | ornithos/lpsvm-master | frankWolfe2.m | .m | lpsvm-master/Optim Ideas/frankWolfe2.m | 3,835 | utf_8 | 05623ed02bf027c3223762561422c4bd | function [u, beta, exitflag] = frankWolfe2(H, initial, D, maxIter, epsilon, ...
tol, searchType, dbg)
% Second Frank-Wolfe (Reverse) solution of the minimax problem for LPSVM.
% Here we have relaxed the problem to a continuous set consisting of all
% convex combinations of H, and used Von Neumann/Sion Theorem to vi... |
github | ornithos/lpsvm-master | SCDNNPrimal2a.m | .m | lpsvm-master/Optim Ideas/SCDNNPrimal2a.m | 2,268 | utf_8 | d7169a8dfe0a1c903838a2e664d792da | function x = SCDNNPrimal2a(f, A, b, lambda, mu, x0, maxIter, tol)
% (Stochastic) Coordinate Ascent for non-negative quadratic minimisation.
% (2) We now actually take analytic (optimal) step for each j rather than
% gradient step. Reduced compute time.
%
% Since we expect a sparse vector, we can reduce the probability ... |
github | ornithos/lpsvm-master | chooseInitColSKM.m | .m | lpsvm-master/Optim Ideas/chooseInitColSKM.m | 2,063 | utf_8 | 2e0a7a7cc90bc1056d60efd2ac022cff |
function I = chooseInitColSKM(H, NN, D, maxIter)
% Perform Spherical KMeans on the remaining columns
% in order to find the most representative subset of the columns
% under scale invariance.
% CURRENTLY IGNORING THE FACT THAT MAY BE CHOOSING MANY SIMILAR TO THE WARM
% START COLUMNS SPECIFIED BEFORE THIS FUNCTION...... |
github | ornithos/lpsvm-master | augLagOpt3.m | .m | lpsvm-master/Optim Ideas/augLagOpt3.m | 2,402 | utf_8 | c88b887c20370341a80c7d7fe650281f | function [x, y] = augLagOpt3(f, A, b, x0, maxIter, tol, lambda, dblA)
% augLagOpt 3rd Augmented Lagrangian Optimisation of the LPSVM Dual Problem
% GULER. CONJUGADTE GRAD VERSION (Alg 2.1)
% This is similar to a Method of Multipliers approach
% lambda are still stationary
% Arguments:
% f - (Col V... |
github | ornithos/lpsvm-master | ADMMDirect4.m | .m | lpsvm-master/Optim Ideas/ADMM/ADMMDirect4.m | 7,695 | utf_8 | 5d2d498d77c768c4c5f2c8da201ef86b | function [primal, dual, fval, exitflag] = ADMMDirect4(H, D, rho, eta, maxIter, ...
tol, reltol, x0, dualreq, verbose)
% ADMMDirect: Solve LP via ADMM using matrix factorisations.
% See Boyd & Parikh chapter 5. The idea is to solve f^T x s.t. A*x = b
% and sum(x) = 1 according to an unconstra... |
github | ornithos/lpsvm-master | ADMMDistributed7.m | .m | lpsvm-master/Optim Ideas/ADMM/ADMMDistributed7.m | 10,868 | utf_8 | a025336b06a76217724e23c87050dab2 | function [primal, dual, fval, exitflag] = ADMMDistributed7(H, D, rho, alpha, ...
maxIter, tol, reltol, x0, npar, dualreq, verbose)
% ADMMDistributed: Solve LP via ADMM using distributed version of
% Boyd & Parikh chapter 5. The idea is to solve f^T x s.t. A*x = b
% and sum(x) = 1 according t... |
github | ornithos/lpsvm-master | ADMMDistributed1.m | .m | lpsvm-master/Optim Ideas/ADMM/ADMMDistributed1.m | 9,808 | utf_8 | a660e0fe2cd87314bc87bf59d4c7181c | function [primal, dual, fval, exitflag] = ADMMDistributed1(H, D, rho, eta, ...
maxIter, tol, reltol, x0, npar, dualreq, verbose)
% ADMMDistributed: Solve LP via ADMM using distributed version of
% Boyd & Parikh chapter 5. The idea is to solve f^T x s.t. A*x = b
% and sum(x) = 1 according to ... |
github | ornithos/lpsvm-master | ADMMDistributed9.m | .m | lpsvm-master/Optim Ideas/ADMM/ADMMDistributed9.m | 12,454 | utf_8 | 3607b2769c9ab3500c36e53c671b9288 | function [primal, fval, exitflag, zu] = ADMMDistributed9(H, D, rho, alpha, ...
maxIter, tol, reltol, npar, x0, search_typ, verbose)
% ADMMDistributed: Solve LP via ADMM using distributed version of
% Boyd & Parikh chapter 5. The idea is to solve f^T x s.t. A*x = b
% and sum(x) = 1 according ... |
github | ornithos/lpsvm-master | lsBoundarySimplex.m | .m | lpsvm-master/Optim Ideas/ADMM/lsBoundarySimplex.m | 1,097 | utf_8 | 71c4770f89950b9a3ca2baf9d72339c3 | function [outA, outB] = lsBoundarySimplex(a, b)
% lsBoundarySimplex(a, b) Shrink/Grow a line segment described by [a,b]
% such that the start and end point are on the boundary of the simplex
% Warning: fminbnd can return Infinite if it fails to find finite values in
% initial iterates. Hence the while loops.
a ... |
github | ornithos/lpsvm-master | ADMMlinear_svm2.m | .m | lpsvm-master/Optim Ideas/ADMM/ADMMlinear_svm2.m | 2,324 | utf_8 | 4f13e57ea2c14562fa8d499c239b9618 | function [xave, history] = linear_svm(A, z, u, rho, alpha)
% linear_svm2
%
% Adapted from Boyd et al. Solves the LPSVM problem on a one-at-a-time
% schedule.
%
% Solves the following problem via ADMM:
%
% minimize (h'*x)_+ - c'*x + (rho/2)||x-z+u||_2^2
% s.t. a \in simplex
% for each i in [n].
%
% This funct... |
github | ornithos/lpsvm-master | ADMMDistributed0.m | .m | lpsvm-master/Optim Ideas/ADMM/ADMMDistributed0.m | 9,901 | utf_8 | 4464e5051ee0a584f97b9e143b56fee6 | function [primal, dual, fval, exitflag] = ADMMDistributed1(H, D, rho, eta, ...
maxIter, tol, reltol, x0, npar, dualreq, verbose)
% ADMMDistributed: Solve LP via ADMM using distributed version of
% Boyd & Parikh chapter 5. The idea is to solve f^T x s.t. A*x = b
% and sum(x) = 1 according to ... |
github | ornithos/lpsvm-master | solve2DBisection2.m | .m | lpsvm-master/Optim Ideas/ADMM/solve2DBisection2.m | 735 | utf_8 | 6371d797146a776c43330112d770d58b | function [a, rho, xi, fval] = solve2DBisection2(H, D, rng_start, rng_end)
% solve2DBisection2(H, D, start, end)
% for general H. We minimise the function
% rhoObj = @(x,a) sum(max(0,x-H'*a)) - x;
% where a is on the line [rng_start, rng_end].
tol = 1e-10; % all evaluations are basically free, so 10^-10 seems g... |
github | ornithos/lpsvm-master | solveLPActiveAD1.m | .m | lpsvm-master/Optim Ideas/ADMM/solveLPActiveAD1.m | 9,363 | utf_8 | 8a7628b09acf51560a5ccadc4687491d | function[out, exitflag, primal, time, msgl] = solveLPActiveAD1(H, D, options, ...
wrm_strt, NN, ass, npar, npiter, tol, dbg, verbose)
% solveLPActiveAD1:
% ----------------- Solve LP via active set identification. Unlike
% solveLPActiveAB1, this method uses the (serial) ADMM optimization,
% which... |
github | ornithos/lpsvm-master | ADMMDirect3.m | .m | lpsvm-master/Optim Ideas/ADMM/ADMMDirect3.m | 4,692 | utf_8 | 8fa8615ed37d0011c662a139dd87d88e | function [z, history] = ADMMDirect3(H, D, rho, alpha, maxIter, ...
tol, reltol, verbose)
% ADMMDirect: Solve LP via ADMM using matrix factorisations.
% See Boyd & Parikh chapter 5. The idea is to solve f^T x s.t. A*x = b
% and sum(x) = 1 according to an unconstrained linear system.
%
% v3... |
github | ornithos/lpsvm-master | solve2DBisection.m | .m | lpsvm-master/Optim Ideas/ADMM/solve2DBisection.m | 657 | utf_8 | a885f1aabec7e9bd94bcdc83949e0427 | function [a, rho, xi, fval] = solve2DBisection(H, D)
% solve2DBisection(H)
% H is a 2 row matrix. We minimise the function
% rhoObj = @(x,j) sum(max(0,x-H'*[j; 1-j])) - x;
[n,~] = size(H);
if n ~=2; error('H must be a 2 column matrix'); end;
tol = 1e-10; % all evaluations are basically free, so 10^-10 seems goo... |
github | ornithos/lpsvm-master | ADMMlinear_svm.m | .m | lpsvm-master/Optim Ideas/ADMM/ADMMlinear_svm.m | 2,977 | utf_8 | f50515f571c885e2e55ef294daff47cf | function [xave, history] = linear_svm(A, lambda, p, rho, alpha)
% linear_svm Solve linear support vector machine (SVM) via ADMM
%
% [x, history] = linear_svm(A, lambda, p, rho, alpha)
%
% Solves the following problem via ADMM:
%
% minimize (1/2)||w||_2^2 + \lambda sum h_j(w, b)
%
% where A is a matrix given by [-... |
github | ornithos/lpsvm-master | ADMMDistributed8.m | .m | lpsvm-master/Optim Ideas/ADMM/ADMMDistributed8.m | 10,194 | utf_8 | 7b2f8c520951b5effbf2a3dd301dc6d4 | function [primal, dual, fval, exitflag] = ADMMDistributed8(H, D, rho, alpha, ...
maxIter, tol, reltol, x0, npar, dualreq, verbose)
% ADMMDistributed: Solve LP via ADMM using distributed version of
% Boyd & Parikh chapter 5. The idea is to solve f^T x s.t. A*x = b
% and sum(x) = 1 according t... |
github | ornithos/lpsvm-master | ADMMlinprogDist2.m | .m | lpsvm-master/Optim Ideas/ADMM/ADMMlinprogDist2.m | 3,494 | utf_8 | bf8d5f2b7fc3bfae9b196b5121fc0d2e | function [z, history] = linprogDist2(c, A, b, ignore, rho, npar, MAX_ITER)
% linprog Solve standard form LP via ADMM
%
% [x, history] = linprog(c, A, b, rho, alpha);
%
% GENERAL CONSENSUS FORM
% Solves the following problem via ADMM:
%
% minimize c'*x
% subject to Ax = b, x >= 0
%
% The solution is returned ... |
github | ornithos/lpsvm-master | ADMMDirect.m | .m | lpsvm-master/Optim Ideas/ADMM/ADMMDirect.m | 3,799 | utf_8 | 5c002facbeee2eafe84fef49e4760c81 | function [z, history] = ADMMDirect(H, D, rho, alpha, maxIter, ...
tol, reltol, verbose)
% ADMMDirect: Solve LP via ADMM using matrix factorisations.
% See Boyd & Parikh chapter 5. The idea is to solve f^T x s.t. A*x = b
% and sum(x) = 1 according to an unconstrained linear system. We then
% ... |
github | ornithos/lpsvm-master | ADMMlinear_svm3.m | .m | lpsvm-master/Optim Ideas/ADMM/ADMMlinear_svm3.m | 2,615 | utf_8 | 5326d709b0e34ac04e50f44ed6a2a1a2 | function [xave, history] = linear_svm3(A, z, u, rho, alpha, tol)
% linear_svm3
%
% -- Using the hinge loss ADMM solver instead of CVX
% Adapted from Boyd et al. Solves the LPSVM problem on a one-at-a-time
% schedule.
%
% Solves the following problem via ADMM:
%
% minimize (h'*x)_+ - c'*x + (rho/2)||x-z+u||_2^2
%... |
github | ornithos/lpsvm-master | ADMMDistributed3.m | .m | lpsvm-master/Optim Ideas/ADMM/ADMMDistributed3.m | 9,604 | utf_8 | d0e3b0a0860177259d28435e7146dc0d | function [primal, dual, fval, exitflag] = ADMMDistributed3(H, D, rho, eta, ...
maxIter, tol, reltol, x0, npar, dualreq, verbose)
% ADMMDistributed: Solve LP via ADMM using distributed version of
% Boyd & Parikh chapter 5. The idea is to solve f^T x s.t. A*x = b
% and sum(x) = 1 according to ... |
github | ornithos/lpsvm-master | ADMMlinprogRplus.m | .m | lpsvm-master/Optim Ideas/ADMM/ADMMlinprogRplus.m | 2,053 | utf_8 | 0685addc169cc25c5e0ab94dc17f54c1 | function [z, history] = linprog(c, A, b, rho, alpha)
% linprog Solve standard form LP via ADMM
%
% [x, history] = linprog(c, A, b, rho, alpha);
%
% Solves the following problem via ADMM:
%
% minimize c'*x
% subject to Ax = b, x >= 0
%
% The solution is returned in the vector x.
%
% history is a structure tha... |
github | ornithos/lpsvm-master | ADMMDistributed6.m | .m | lpsvm-master/Optim Ideas/ADMM/ADMMDistributed6.m | 11,317 | utf_8 | 9a55711c1e02ef3a765b3f5ae4641984 | function [primal, dual, fval, exitflag] = ADMMDistributed6(H, D, rho, eta, ...
maxIter, tol, reltol, x0, npar, dualreq, verbose)
% ADMMDistributed: Solve LP via ADMM using distributed version of
% Boyd & Parikh chapter 5. The idea is to solve f^T x s.t. A*x = b
% and sum(x) = 1 according to ... |
github | ornithos/lpsvm-master | ADMMDistributed5.m | .m | lpsvm-master/Optim Ideas/ADMM/ADMMDistributed5.m | 12,268 | utf_8 | 554a24a18bbef41c352dc9a2a20afc24 | function [primal, dual, fval, exitflag] = ADMMDistributed5(H, D, rho, eta, ...
maxIter, tol, reltol, npar, force_ls, dualreq, verbose)
% ADMMDistributed: Solve LP via ADMM using distributed version of
% Boyd & Parikh chapter 5. The idea is to solve f^T x s.t. A*x = b
% and sum(x) = 1 accordi... |
github | ornithos/lpsvm-master | ADMMlinprogDist.m | .m | lpsvm-master/Optim Ideas/ADMM/ADMMlinprogDist.m | 3,168 | utf_8 | ff397f7fa856209aaf97e3106a4bd454 | function [z, history] = linprogDist(c, A, b, rho, npar)
% linprog Solve standard form LP via ADMM
%
% [x, history] = linprog(c, A, b, rho, alpha);
%
% Solves the following problem via ADMM:
%
% minimize c'*x
% subject to Ax = b, x >= 0
%
% The solution is returned in the vector x.
%
% history is a structure ... |
github | ornithos/lpsvm-master | ADMMDistributed4.m | .m | lpsvm-master/Optim Ideas/ADMM/ADMMDistributed4.m | 10,102 | utf_8 | 356dd069a7e7f153627139943f7bd419 | function [primal, dual, fval, exitflag] = ADMMDistributed4(H, D, rho, eta, ...
maxIter, tol, reltol, x0, npar, dualreq, verbose)
% ADMMDistributed: Solve LP via ADMM using distributed version of
% Boyd & Parikh chapter 5. The idea is to solve f^T x s.t. A*x = b
% and sum(x) = 1 according to ... |
github | ornithos/lpsvm-master | ADMMDistributed2.m | .m | lpsvm-master/Optim Ideas/ADMM/ADMMDistributed2.m | 9,699 | utf_8 | a56b4733e03b9d22df271539380b57af | function [primal, dual, fval, exitflag] = ADMMDistributed2(H, D, rho, eta, ...
maxIter, tol, reltol, x0, npar, dualreq, verbose)
% ADMMDistributed: Solve LP via ADMM using distributed version of
% Boyd & Parikh chapter 5. The idea is to solve f^T x s.t. A*x = b
% and sum(x) = 1 according to ... |
github | ornithos/lpsvm-master | solveLPActiveABtest.m | .m | lpsvm-master/Testing and Auxiliary/solveLPActiveABtest.m | 6,723 | utf_8 | 11d3c1472358b196de87187d16c4ab72 | function [tt,betahist,sz,nz] = solveLPActiveABtest(H, D, options, wrm_strt, NN, ass, dbg)
%SOLVELPACTIVE Solve LP with active set strategy
tol = 1e-5;
time = zeros(1, 200);
% Initialise size vars
[dim, N] = size(H);
if NN < 1; NN = ceil(N*NN); end;
if NN == 0; NN = ceil(N/100); end;
if ceil(1/D) + dim < N
num_ne... |
github | YuvalNirkin/DenseCorrespondences-master | findTexturelessRegions.m | .m | DenseCorrespondences-master/Matlab/findTexturelessRegions.m | 3,547 | utf_8 | d8c5be4c3f958492904aeafca3288b92 | % *************************************************************************
% Title: Function-Find Textureless regions of an image
% Notes: Textureless regions are defined as regions where the squared horizontal
% intensity gradient averaged over a square window of a given size
% (windowSize) is below a given threshol... |
github | YuvalNirkin/DenseCorrespondences-master | findTexturelessRegions2.m | .m | DenseCorrespondences-master/Matlab/findTexturelessRegions2.m | 3,636 | utf_8 | 24d0d8144a6be6f1c31a14ad3ac91dc5 | % *************************************************************************
% Title: Function-Find Textureless regions of an image
% Notes: Textureless regions are defined as regions where the squared horizontal
% intensity gradient averaged over a square window of a given size
% (windowSize) is below a given threshol... |
github | YuvalNirkin/DenseCorrespondences-master | get_descriptors.m | .m | DenseCorrespondences-master/Matlab/scalemaps/get_descriptors.m | 3,348 | utf_8 | 401362db8a10ce58b80f279f7506ce97 | function [desc1, desc2] = get_descriptors(image1, image2, proptype, weight_func)
% Extract dense descriptors [1] for an image pair. These can then be used
% to establish dense correspondences between the images using the SIFT-Flow
% algorithm [2].
%
% Project webpage:
% http://www.openu.ac.il/home/hassner/scalemaps
... |
github | YuvalNirkin/DenseCorrespondences-master | propagateScales.m | .m | DenseCorrespondences-master/Matlab/scalemaps/propagateScales.m | 7,899 | utf_8 | f48b0bf4dd3611679ba28f3070926ae8 | function [varargout] = propagateScales(varargin)
% This method is used for propagating scales from sparse interest points
% to neighboring pixels.
% This method contains three different means for scale propagation:
% (1) using only the scales at detected interest points ('geometric'),
% (2) using the underlying image i... |
github | YuvalNirkin/DenseCorrespondences-master | warpFL.m | .m | DenseCorrespondences-master/Matlab/Tests/sift_flow_test/warpFL.m | 212 | utf_8 | 8a88e59d40dc4a442477f98d01b9301b | % warp i2 according to flow field in vx vy
function [warpI2,I]=warp(i2,vx,vy)
[M,N]=size(i2);
[x,y]=meshgrid(1:N,1:M);
warpI2=interp2(x,y,i2,x+vx,y+vy,'bicubic');
I=find(isnan(warpI2));
warpI2(I)=zeros(size(I));
|
github | YuvalNirkin/DenseCorrespondences-master | computeColor.m | .m | DenseCorrespondences-master/Matlab/Tests/sift_flow_test/computeColor.m | 3,142 | utf_8 | a36a650437bc93d4d8ffe079fe712901 | function img = computeColor(u,v)
% computeColor color codes flow field U, V
% According to the c++ source code of Daniel Scharstein
% Contact: schar@middlebury.edu
% Author: Deqing Sun, Department of Computer Science, Brown University
% Contact: dqsun@cs.brown.edu
% $Date: 2007-10-31 21:20:30 (Wed, 31 O... |
github | YuvalNirkin/DenseCorrespondences-master | warpImage.m | .m | DenseCorrespondences-master/Matlab/Tests/sift_flow_test/warpImage.m | 481 | utf_8 | fc40d048af1746aef3b59b2ec6fb50a7 | % function to warp images with different dimensions
function [warpI2,mask]=warpImage(im,vx,vy)
[height2,width2,nchannels]=size(im);
[height1,width1]=size(vx);
[xx,yy]=meshgrid(1:width2,1:height2);
[XX,YY]=meshgrid(1:width1,1:height1);
XX=XX+vx;
YY=YY+vy;
mask=XX<1 | XX>width2 | YY<1 | YY>height2;
XX=min(max(XX,1),wid... |
github | YuvalNirkin/DenseCorrespondences-master | warpFLColor.m | .m | DenseCorrespondences-master/Matlab/Tests/sift_flow_test/warpFLColor.m | 495 | utf_8 | b579d13197cea640cfb9603adf463a67 | % Function to warp color image im2 to the grid of im1. It uses the pixels
% in im1 to fill in the holes of warpI2 if there is any in the warping
function warpI2=warpFLColor(im1,im2,vx,vy)
if isfloat(im1)~=1
im1=im2double(im1);
end
if isfloat(im2)~=1
im2=im2double(im2);
end
if exist('vy')~=1
vy=vx(:,:,2);
... |
github | YuvalNirkin/DenseCorrespondences-master | get_descriptors_test.m | .m | DenseCorrespondences-master/Matlab/Tests/do_propagateScales/get_descriptors_test.m | 3,378 | utf_8 | 51200d99a082c8e8caed398645e4303d | function [desc1, desc2] = get_descriptors_test(image1, image2, proptype, weight_func)
% Extract dense descriptors [1] for an image pair. These can then be used
% to establish dense correspondences between the images using the SIFT-Flow
% algorithm [2].
%
% Project webpage:
% http://www.openu.ac.il/home/hassner/scale... |
github | YuvalNirkin/DenseCorrespondences-master | propagateScales_test.m | .m | DenseCorrespondences-master/Matlab/Tests/do_propagateScales/propagateScales_test.m | 7,909 | utf_8 | cf5097e43ee5442bb9f8b6894ca304ba | function [varargout] = propagateScales_test(varargin)
% This method is used for propagating scales from sparse interest points
% to neighboring pixels.
% This method contains three different means for scale propagation:
% (1) using only the scales at detected interest points ('geometric'),
% (2) using the underlying im... |
github | YuvalNirkin/DenseCorrespondences-master | warpFL.m | .m | DenseCorrespondences-master/Matlab/SIFTflow/warpFL.m | 212 | utf_8 | 8a88e59d40dc4a442477f98d01b9301b | % warp i2 according to flow field in vx vy
function [warpI2,I]=warp(i2,vx,vy)
[M,N]=size(i2);
[x,y]=meshgrid(1:N,1:M);
warpI2=interp2(x,y,i2,x+vx,y+vy,'bicubic');
I=find(isnan(warpI2));
warpI2(I)=zeros(size(I));
|
github | YuvalNirkin/DenseCorrespondences-master | computeColor.m | .m | DenseCorrespondences-master/Matlab/SIFTflow/computeColor.m | 3,142 | utf_8 | a36a650437bc93d4d8ffe079fe712901 | function img = computeColor(u,v)
% computeColor color codes flow field U, V
% According to the c++ source code of Daniel Scharstein
% Contact: schar@middlebury.edu
% Author: Deqing Sun, Department of Computer Science, Brown University
% Contact: dqsun@cs.brown.edu
% $Date: 2007-10-31 21:20:30 (Wed, 31 O... |
github | YuvalNirkin/DenseCorrespondences-master | SIFTflowc2f.m | .m | DenseCorrespondences-master/Matlab/SIFTflow/SIFTflowc2f.m | 5,485 | utf_8 | a5d8f4d01080d208613afeb9cce2e799 | % function to do coarse to fine SIFT flow matching
function [vx,vy,energylist]=SIFTflowc2f(im1,im2,SIFTflowpara,isdisplay,Segmentation)
if isfield(SIFTflowpara,'alpha')
alpha=SIFTflowpara.alpha;
else
alpha=0.01;
end
if isfield(SIFTflowpara,'d')
d=SIFTflowpara.d;
else
d=alpha*20;
end
if isfield(SIFTfl... |
github | YuvalNirkin/DenseCorrespondences-master | warpImage.m | .m | DenseCorrespondences-master/Matlab/SIFTflow/warpImage.m | 481 | utf_8 | fc40d048af1746aef3b59b2ec6fb50a7 | % function to warp images with different dimensions
function [warpI2,mask]=warpImage(im,vx,vy)
[height2,width2,nchannels]=size(im);
[height1,width1]=size(vx);
[xx,yy]=meshgrid(1:width2,1:height2);
[XX,YY]=meshgrid(1:width1,1:height1);
XX=XX+vx;
YY=YY+vy;
mask=XX<1 | XX>width2 | YY<1 | YY>height2;
XX=min(max(XX,1),wid... |
github | YuvalNirkin/DenseCorrespondences-master | warpFLColor.m | .m | DenseCorrespondences-master/Matlab/SIFTflow/warpFLColor.m | 495 | utf_8 | b579d13197cea640cfb9603adf463a67 | % Function to warp color image im2 to the grid of im1. It uses the pixels
% in im1 to fill in the holes of warpI2 if there is any in the warping
function warpI2=warpFLColor(im1,im2,vx,vy)
if isfloat(im1)~=1
im1=im2double(im1);
end
if isfloat(im2)~=1
im2=im2double(im2);
end
if exist('vy')~=1
vy=vx(:,:,2);
... |
github | ppbits/vqa-master | SteerableFeatureSetProperties.m | .m | vqa-master/MATLAB/SteerableFeatureSetProperties.m | 9,324 | utf_8 | e0f45110f97bf0444523b0227a0deb68 | %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% ---
% Written by: Konstantinos G. Derpanis
% PhD Candidate
% York University
% Toronto, Ontario, Canada
% Bugs/Questions email me at: kosta@cs.yorku.ca
% see York Universtiy Technical Report CS-2004-05 for analytic det... |
github | ppbits/vqa-master | WriteLog.m | .m | vqa-master/MATLAB/WriteLog.m | 465 | utf_8 | 3f755b5b3414e52bf46b0bca450b6cdb | %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% Writes a log
%
% Name: Peng Peng
% Contact: dante.peng@gmail.com
% Date: Sept 20, 2015
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
function WriteLog(str)
timesta... |
github | ppbits/vqa-master | ComputeSpatialQuality.m | .m | vqa-master/MATLAB/ComputeSpatialQuality.m | 609 | utf_8 | adc6c12cdda567e5b55ae7d3625ab524 | %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% Computes the spatial quality of the video using the MS-SSIM index.
%
% Name: Peng Peng
% Contact: dante.peng@gmail.com
% Date: Oct 19, 2015
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%... |
github | ppbits/vqa-master | GetQualityScores.m | .m | vqa-master/MATLAB/GetQualityScores.m | 1,574 | utf_8 | b9e894ab484dbc29a4e5782b1a4d6e54 | %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% Get quality scores
%
% Reads the scores from disk if they are already computed and stored. Otherwise, compute the
% quality scores by calling the ComputeQualtyHandle.
%
% Name: Peng Peng
% Contact: dante.peng@gmail.co... |
github | ppbits/vqa-master | ComputeCenterBias.m | .m | vqa-master/MATLAB/ComputeCenterBias.m | 559 | utf_8 | b3123583f50eb55ef81df3f3cfcd0e3d | %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% Compute center bias map for visual attention
%
% Name: Peng Peng
% Contact: dante.peng@gmail.com
% Date: Oct 11, 2015
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
... |
github | ppbits/vqa-master | MeasureDistortion.m | .m | vqa-master/MATLAB/MeasureDistortion.m | 2,730 | utf_8 | 10f18e58696ffa0ff56e934e24de19e0 | %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% This function computes the saliency weighted and motion tuned distortion metric
% between two volumes of spatiotemporal energy distributions.
%
% Name: Peng Peng
% Contact: dante.peng@gmail.com
% Date: Sept 20, 2015
%%... |
github | ppbits/vqa-master | TemporalPooling.m | .m | vqa-master/MATLAB/TemporalPooling.m | 1,027 | utf_8 | 8c5efff0b2995cd2aa7c22b0f78fb60c | %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% Temporal Pooling Method
%
% Name: Peng Peng
% Contact: dante.peng@gmail.com
% Date: Nov 17, 2015
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
function dv = Tempora... |
github | ppbits/vqa-master | SpatiotemporalOrientationAnalysis.m | .m | vqa-master/MATLAB/SpatiotemporalOrientationAnalysis.m | 20,848 | utf_8 | a720c7a473117e4765dafc0bd686518b | %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% ---
% Written by: Konstantinos G. Derpanis
% PhD Candidate
% York University
% Toronto, Ontario, Canada
% Bugs/Questions email me at: kosta@cs.yorku.ca
% see York Universtiy Technical Report CS-2004-05 for analytic det... |
github | ppbits/vqa-master | WriteResultLine.m | .m | vqa-master/MATLAB/WriteResultLine.m | 492 | utf_8 | 88977a8f262859620c6fac1a64d2a250 | %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% Writes to the 'Results.txt' file.
%
% Name: Peng Peng
% Contact: dante.peng@gmail.com
% Date: Oct 17, 2015
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
function Wr... |
github | ppbits/vqa-master | GetDistortionTypeName.m | .m | vqa-master/MATLAB/GetDistortionTypeName.m | 1,152 | utf_8 | 163e554711092d2c48faff1d1a99fa7a | %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% Gets the distortion type name from the distortin type index
%
% Name: Peng Peng
% Contact: dante.peng@gmail.com
% Date: Oct 17, 2015
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%... |
github | ppbits/vqa-master | DownSample.m | .m | vqa-master/MATLAB/DownSample.m | 733 | utf_8 | 69605a2d4c40e9f5512965ea98e62638 | %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% Down-samples a video
%
% Name: Peng Peng
% Contact: dante.peng@gmail.com
% Date: Sept 20, 2015
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
function v2 = DownSampl... |
github | ppbits/vqa-master | Logistic.m | .m | vqa-master/MATLAB/Logistic.m | 554 | utf_8 | adbebba813aa786fe2db2ffefcf70985 | %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% Performs logistic mapping
%
% Name: Peng Peng
% Contact: dante.peng@gmail.com
% Date: Sept 20, 2015
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
function yhat = Lo... |
github | ppbits/vqa-master | ComputeMotionQuality.m | .m | vqa-master/MATLAB/ComputeMotionQuality.m | 2,690 | utf_8 | 4f2fbf0a8886856e2f1457f1e925f5e3 | %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% Compute Attention Guided and Motion Tuned Quality based on Spatiotemporal Orientation Analysis
%
% Name: Peng Peng
% Contact: dante.peng@gmail.com
% Date: Sept 20, 2015
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%... |
github | ppbits/vqa-master | TestLiveMobile.m | .m | vqa-master/MATLAB/TestLiveMobile.m | 4,849 | utf_8 | 04244477c811452432f25cf3029bab0c | %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% Evaluate on the LIVE Mobile VQA Database
%
% Name: Peng Peng
% Contact: dante.peng@gmail.com
% Date: Oct 11, 2015
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
func... |
github | ppbits/vqa-master | WriteResult.m | .m | vqa-master/MATLAB/WriteResult.m | 464 | utf_8 | 2d0d456448ac8954b5a0a9f0a9721043 | %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% Writes to the 'Results.txt' file.
%
% Name: Peng Peng
% Contact: dante.peng@gmail.com
% Date: Oct 17, 2015
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
function Wr... |
github | ppbits/vqa-master | ReadVideo.m | .m | vqa-master/MATLAB/ReadVideo.m | 1,365 | utf_8 | eb725bcd69e740549a42b44ddeaba49d | %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% Reads a video at a specific scale.
%
% If the video has already been scaled and stored on the disk as .mat file, read in the .mat file.
% Otherwise, reads in the .yuv file, down-sample it to the desired scale and stor... |
github | ppbits/vqa-master | TestLive.m | .m | vqa-master/MATLAB/TestLive.m | 4,564 | utf_8 | cd66a4ac4b3f0ed7248fa420fa4cb6dc | %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% Evaluate on the LIVE VQA Database
%
% Name: Peng Peng
% Contact: dante.peng@gmail.com
% Date: Oct 18, 2015
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
function Te... |
github | ppbits/vqa-master | GetSaliency.m | .m | vqa-master/MATLAB/GetSaliency.m | 626 | utf_8 | d29308d404438120a77558cdc0c7ff0f | %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% Gets the saliency map based on self-information.
%
% Name: Peng Peng
% Contact: dante.peng@gmail.com
% Date: Sept 20, 2015
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%... |
github | ppbits/vqa-master | Test.m | .m | vqa-master/MATLAB/Test.m | 852 | utf_8 | 7dd9b14595efdcbdead2b61f79fd3da0 | %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% Evaluate the proposed VQA algorithm
%
% Name: Peng Peng
% Contact: dante.peng@gmail.com
% Date: Oct 17, 2015
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
function ... |
github | ppbits/vqa-master | GetDistortionTypes.m | .m | vqa-master/MATLAB/GetDistortionTypes.m | 1,703 | utf_8 | 5a9cc6b774ab78c1591d708caab35065 | %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% Gets a vector representing the distortion types of all distorted videos in a database
%
% Name: Peng Peng
% Contact: dante.peng@gmail.com
% Date: Sept 20, 2015
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%... |
github | ppbits/vqa-master | Performance.m | .m | vqa-master/MATLAB/Performance.m | 2,063 | utf_8 | 58e86840d8eb4b8157a958b5d9daa770 | %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% Gets the correlation between the predicted quality and subject quality.
%
% 4 metrics included:
% (1) SRCC (Spearman's Rank Correlation Coefficient)
% (1) KRCC (Kendall's Rank Correlation Coefficient)
% (1... |
github | ppbits/vqa-master | ComputeSelfInformation.m | .m | vqa-master/MATLAB/ComputeSelfInformation.m | 1,588 | utf_8 | 5a339c5c8ae7e3d581fddfbec6b621e2 | %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% Computes self-information
%
% Name: Peng Peng
% Contact: dante.peng@gmail.com
% Date: Sept 20, 2015
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
function volSI = C... |
github | ppbits/vqa-master | ReadYUV.m | .m | vqa-master/MATLAB/ReadYUV.m | 1,872 | utf_8 | 8377e2f79d7bb15260d9751fadf220b5 | %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% (c) Mani Malek Esmaeili
% This function reads a YUV file assuming that the file into RGB matrices
% of the same size (as frame_size)
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%... |
github | ppbits/vqa-master | ApplyCenterBias.m | .m | vqa-master/MATLAB/ApplyCenterBias.m | 501 | utf_8 | 6559191161784eeac2bebd6f24eec991 | %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% Apply center bias to a saliency map
%
% Name: Peng Peng
% Contact: dante.peng@gmail.com
% Date: Sept 20, 2015
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
function... |
github | ppbits/vqa-master | CombineSaliency.m | .m | vqa-master/MATLAB/CombineSaliency.m | 638 | utf_8 | e267aad9e437bdd28d99d982abc2e7b7 | %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% Combine two saliency maps
%
% Name: Peng Peng
% Contact: dante.peng@gmail.com
% Date: Oct 11, 2015
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
function s = Combin... |
github | kylemcdonald/structured-light-master | structuredlight.m | .m | structured-light-master/Matlab/structuredlight.m | 7,633 | utf_8 | fb09a867eb054d163e5d43fcaeba7b88 | function [t,angles,unwrapped_angles,p1,p2,p3,pd]=structuredlight(p1name,p2name,p3name,gamma,rotate,hgtname);
% binarymillenium Jan 2010
% GNU GPL v3.0
% phases generates any number of equally spaced sine waves.
%
% slight (need to rename to avoid collision with existing matlab slight) loads images and performs ... |
github | kylemcdonald/structured-light-master | phases.m | .m | structured-light-master/Matlab/phases.m | 562 | utf_8 | eb44992b98cb375dff32f33d7194f1d6 | %%
function structured_light()
gen_sawtooths(3);
function gen_sawtooths(n)
t = [1:500];
y = zeros(n, length(t));
for i = [1:n]
new_y = sawtooth(t,floor(length(t)/n)*(i-1)+1);
y(i,:) = new_y;
% y2 = sawtooth(t,floor(500/n));
% y3 = sawtooth(t,floor(500/3*2));
end
figure(1);
... |
github | chenggiant/dump1090-matlab-master | decode_adsb_signal.m | .m | dump1090-matlab-master/decode_adsb_signal.m | 11,934 | utf_8 | 635fd9774a792cb5c96673182bec9a56 | function [packetDataCell] = decode_adsb_signal(skipSeconds, filename, dataPath,step,sampRate)
nbrSamples = step * sampRate;
q = ceil(sampRate/1e7*10);
fc = 1090e6;
Ts = 1/sampRate;
% Loading data
fprintf(sprintf(' -- Loading data ... \n'));
rxSignal = read_binary(sprintf('%s/%s',dataPat... |
github | chenggiant/dump1090-matlab-master | decode_cpr.m | .m | dump1090-matlab-master/decode_cpr.m | 4,570 | utf_8 | 81f2fd0622d534ad049909731d36054a | function [a] = decode_cpr(a)
AirDlat0 = 360.0 / 60;
AirDlat1 = 360.0 / 59;
lat0 = a.even_cprlat;
lat1 = a.odd_cprlat;
lon0 = a.even_cprlon;
lon1 = a.odd_cprlon;
% Compute the Latitude Index j
j = floor(((59*lat0 - 60*lat1) / 131072) + 0.5);
rlat0 = AirDlat0 * (cprModFunction(j,60) ... |
github | chenggiant/dump1090-matlab-master | dump1090.m | .m | dump1090-matlab-master/dump1090.m | 7,603 | utf_8 | 9be376d6375820d7565a8cd5a8dd0c54 | % Main function for decoding ADSB, save decoded info in decodedDataCell
function [decodedDataCell] = dump1090(seconds,dataPath,filename,sampRate)
% seconds = 2;
% dataPath = 'D:\USRP_data\ADSB';
% filename = 'adsb_feb13_id5_rx1_set3.dat';
% sampRate = 0.4e7;
step = 0.5; % processing 0... |
github | CentroEPiaggio/HisTOOLogy-master | gCMYK.m | .m | HisTOOLogy-master/src/gCMYK.m | 25,502 | utf_8 | ac12137fa80904a36163bef37facf82b | function varargout = gCMYK(varargin)
% GCMYK MATLAB code for gCMYK.fig
% GCMYK, by itself, creates a new GCMYK or raises the existing
% singleton*.
%
% H = GCMYK returns the handle to a new GCMYK or the handle to
% the existing singleton*.
%
% GCMYK('CALLBACK',hObject,eventData,handles,...) cal... |
github | CentroEPiaggio/HisTOOLogy-master | gRGB.m | .m | HisTOOLogy-master/src/gRGB.m | 20,967 | utf_8 | 61bc6e791eb7fd90ff87fde511acc9ec | function varargout = gRGB(varargin)
% GRGB MATLAB code for gRGB.fig
% GRGB, by itself, creates a new GRGB or raises the existing
% singleton*.
%
% H = GRGB returns the handle to a new GRGB or the handle to
% the existing singleton*.
%
% GRGB('CALLBACK',hObject,eventData,handles,...) calls the l... |
github | CentroEPiaggio/HisTOOLogy-master | HisTOOLogy_index.m | .m | HisTOOLogy-master/src/HisTOOLogy_index.m | 10,870 | utf_8 | 9daea1f36338fdae5e03a0885b5f296a | function varargout = HisTOOLogy_index(varargin)
% HISTOOLOGY_INDEX MATLAB code for HisTOOLogy_index.fig
% HISTOOLOGY_INDEX, by itself, creates a new HISTOOLOGY_INDEX or raises the existing
% singleton*.
%
% H = HISTOOLOGY_INDEX returns the handle to a new HISTOOLOGY_INDEX or the handle to
% the exis... |
github | ismrmrd/ismrmrd-paper-master | do_spiral_recon_matlab.m | .m | ismrmrd-paper-master/code/do_spiral_recon_matlab.m | 5,838 | utf_8 | 1d740e1e6a88512f2fce32637a643cbb | %
% For this code to work, you must add <ISMRMRD_SOURCE>/matlab to your path
% You also need to download the test data
% https://github.com/ismrmrd/ismrmrd/releases/download/v1.2.3-data/ismrmrd_data.zip
% This code requires:
% - vds.m for calculation of spiral trajectories (download from:
% http://www-mrsrl.s... |
github | ismrmrd/ismrmrd-paper-master | do_recon_matlab.m | .m | ismrmrd-paper-master/code/do_recon_matlab.m | 5,342 | utf_8 | f64d86187aae84495425bb4f03bc141d | %
% For this code to work, you must add <ISMRMRD_SOURCE>/matlab to your path
% You also need to download the test data (see do_get_data.sh)
%
function do_recon_matlab()
fnames = {'synth.h5', 'bruker.h5', 'ge.h5', 'philips.h5', 'siemens.h5'};
for p = 1:length(fnames)
recon_dataset(fnames{p});
end
exit
end
%-... |
github | ismrmrd/ismrmrd-paper-master | plotgradinfo.m | .m | ismrmrd-paper-master/code/extern/vdspiral/plotgradinfo.m | 3,685 | utf_8 | 5782aa4aba762b863bfa8a8840e06ffe | %
% function [k,g,s,m1,m2,t,v]=plotgradinfo(g,T,k0)
%
% Function makes a nice plot given a 2D gradient waveform.
%
% INPUT:
% g gradient (G/cm) = gx + i*gy, or Nx2, Nx3 or Nx4.
% T sample period (s)
% k0 initial condition for k-space.
%
% OUTPUT:
% k k-space trajectory (cm^(-1))
% g gradient (G/cm)
% s slew rate ... |
github | ismrmrd/ismrmrd-paper-master | vds.m | .m | ismrmrd-paper-master/code/extern/vdspiral/vds.m | 9,277 | utf_8 | e82fe8bbe2c989599c4a24a65fd11b1b | %
% function [k,g,s,time,r,theta] = vds(smax,gmax,T,N,Fcoeff,rmax)
%
%
% VARIABLE DENSITY SPIRAL GENERATION:
% ----------------------------------
%
% Function generates variable density spiral which traces
% out the trajectory
%
% k(t) = r(t) exp(i*q(t)), [1]
%
% Where q is the same as theta...
% r and q are ... |
github | ismrmrd/ismrmrd-paper-master | calcgradinfo.m | .m | ismrmrd-paper-master/code/extern/vdspiral/calcgradinfo.m | 1,595 | utf_8 | e888e10d970046f47949765dcd3ecee4 | %
% function [k,g,s,m1,m2,t,v]=calcgradinfo(g,T,k0,R,L,eta)
%
% Function calculates gradient information
%
% INPUT:
% g gradient (G/cm) = gx + i*gy
% T sample period (s)
% k0 initial condition for k-space.
% R coil resistance (ohms, default =.35)
% L coil inductance (H, default = .0014)
% eta coil efficiency (G/c... |
github | ismrmrd/ismrmrd-paper-master | vdsmex.m | .m | ismrmrd-paper-master/code/extern/vdspiral/vdsmex.m | 2,704 | utf_8 | bbd4fa107499599235dd07af31eb49dd | %
% function [k,g,s,time] = vdsmex(N,FOV,res,gmax,smax,T,ngmax)
%
%
% VARIABLE DENSITY SPIRAL GENERATION:
% ----------------------------------
% This function uses a mex file to quickly generate a
% variable-density spiral. See vds.m for more details.
%
% INPUT:
% N Number of interleaves.
% FOV Field-of-View* (cm)... |
github | ismrmrd/ismrmrd-paper-master | q2r21.m | .m | ismrmrd-paper-master/code/extern/vdspiral/q2r21.m | 3,029 | utf_8 | 2ae6084816deb09872ed08347c101a93 | %
% function [q2,r2] = q2r2(smax,gmax,r,r1,T,Ts,N,F)
%
% VARIABLE DENSITY SPIRAL DESIGN ITERATION
% ----------------------------------------
% Calculates the second derivative of r and q (theta),
% the slew-limited or FOV-limited
% r(t) and q(t) waveforms such that
%
% k(t) = r(t) exp(i*q(t))
%
% Where the FOV is a... |
github | ismrmrd/ismrmrd-paper-master | mri_r2_fit.m | .m | ismrmrd-paper-master/code/extern/irt/mri/mri_r2_fit.m | 2,320 | utf_8 | 89304fd84ed299f366a4112e9cd4a4e2 | function r2 = mri_r2_fit(images, telist, varargin)
%|function r2 = mri_r2_fit(images, telist, varargin)
%|
%| fit R2=1/T2 maps to images with different echo times
%| in
%| images [(nd) nt] images with nt different echo times
%| telist [nt] echo times
%|
%| options
%| 'how' char 'log-ls' - ordinary LS fit to log dat... |
github | ismrmrd/ismrmrd-paper-master | mri_brainweb_params.m | .m | ismrmrd-paper-master/code/extern/irt/mri/mri_brainweb_params.m | 5,068 | utf_8 | 322c09c10026274a49cfacfeb541c93e | function f = mri_brainweb_params(label, varargin)
%function f = mri_brainweb_params(label, varargin)
%|
%| Tissue parameters for BrainWeb synthetic brain images:
%| 1.5T values from BrainWeb database:
%| http://mouldy.bic.mni.mcgill.ca/brainweb/tissue_mr_parameters.txt
%| 3.0T T1, T2 values taken from literature when ... |
github | ismrmrd/ismrmrd-paper-master | mri_objects.m | .m | ismrmrd-paper-master/code/extern/irt/mri/mri_objects.m | 14,386 | utf_8 | 4e9619f7fe2f72c5f4b033e40a2007cc | function st = mri_objects(varargin)
%|function st = mri_objects([options], 'type1', params1, 'type2', params2, ...)
%| Generate strum that describes image-domain objects and Fourier domain spectra
%| of simple structures such as rectangles, disks, superpositions thereof.
%| These functions are useful for simple "idea... |
github | ismrmrd/ismrmrd-paper-master | mri_kspace_spiral.m | .m | ismrmrd-paper-master/code/extern/irt/mri/mri_kspace_spiral.m | 7,460 | utf_8 | 11f16b4459a7d43315adcd0f8e56e5bc | function [kspace, omega] = mri_kspace_spiral(varargin)
%function [kspace, omega] = mri_kspace_spiral(varargin)
% k-space spiral trajectory based on GE 3T scanner constraints
% options (name / value pairs)
% N size of reconstructed image
% Nt # of time points
% fov field of view in cm
% dt time sampling interval
% out
... |
github | ismrmrd/ismrmrd-paper-master | mri_grid_linear.m | .m | ismrmrd-paper-master/code/extern/irt/mri/mri_grid_linear.m | 2,269 | utf_8 | 28a4e2a24f45655c70d16f2332a74a63 | function [xhat, yhat, xg, kg] = mri_grid_linear(kspace, ydata, N, fov)
%|function [xhat, yhat, xg, kg] = mri_grid_linear(kspace, ydata, N, fov)
%| very crude "gridding" based on linear interpolation.
%| not recommended as anything but a straw man or perhaps
%| for initializing iterative methods.
%| in
%| kspace [M 2... |
github | ismrmrd/ismrmrd-paper-master | mri_sensemap_sim.m | .m | ismrmrd-paper-master/code/extern/irt/mri/mri_sensemap_sim.m | 8,400 | utf_8 | 493d13f0b9a1e50622566e9e0e88f148 | function [smap x y] = mri_sensemap_sim(varargin)
%function [smap x y] = mri_sensemap_sim(varargin)
%|
%| Simulate sensitivity maps for sensitivity-encoded MRI
%| based grivich:00:tmf doi:10.1119/1.19461
%|
%| option
%| nx, ny, dx, dy, ncoil, rcoil, coil_distance, orbit (see below)
%|
%| out
%| smap [nx ny ncoil] simul... |
github | ismrmrd/ismrmrd-paper-master | mri_b1map_sliceselect.m | .m | ismrmrd-paper-master/code/extern/irt/mri/mri_b1map_sliceselect.m | 67,545 | utf_8 | 5005c846f429b6713fae8ad1f159cc95 | function [zmaps omaps cost] = mri_b1map_sliceselect(yy, alpha, varargin)
% function [zmaps omaps cost] = mri_b1map_sliceselect(yy, alpha, [options])
%
% Estimate "B1+ map" for each of ncoil coils
% from sequence of reconstructed images with ntip different nominal tip angles.
% Model:
% todo
% in
% yy [nx ny nmeasure... |
github | ismrmrd/ismrmrd-paper-master | ir_mri_dyn_data_share.m | .m | ismrmrd-paper-master/code/extern/irt/mri/ir_mri_dyn_data_share.m | 4,220 | utf_8 | c480f8b3e4dd503e76c29cb28fefabc8 | function full_data = ir_mri_dyn_data_share(undersamp_data, sampling_pattern)
%function full_data = ir_mri_dyn_data_share(undersamp_data, sampling_pattern)
%|
%| A data-sharing approach to imputing missing k-space data in dynamic MRI.
%| Takes undersampled dynamic data and uses data-sharing (0th order interp)
%| to fil... |
github | ismrmrd/ismrmrd-paper-master | mri_exp_approx.m | .m | ismrmrd-paper-master/code/extern/irt/mri/mri_exp_approx.m | 9,164 | utf_8 | 27c92ffff860b0fc79108efbec53cccb | function [B, C, hk, zk] = mri_exp_approx(ti, zmap, LL, varargin)
%|function [B, C, hk, zk] = mri_exp_approx(ti, zmap, LL, [options])
%|
%| Build approximations to exponentials for iterative MR image reconstruction,
%| generalizing "time segmentation" and "frequency segmentation" methods.
%| This is a key part of the ... |
github | ismrmrd/ismrmrd-paper-master | ir_mri_dce_samp1.m | .m | ismrmrd-paper-master/code/extern/irt/mri/ir_mri_dce_samp1.m | 3,572 | utf_8 | 8452606125252fecc9ab09d2e41a9ecc | function [samp1 samp2] = ir_mri_dce_samp1(dims, varargin)
%function [samp1 samp2] = ir_mri_dce_samp1(dims, varargin)
%|
%| Generate dynamic kspace data for DCE MRI simulations.
%|
%| in
%| dims dynamic object dims, e.g. [nx ny Nt]
%| Nt is # of "phase encode groups"
%|
%| option
%| 'pattern' (char) (default: 'v01... |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.