1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303
|
function ApriltagsDemo(camId, colordepth, tagSize, tagFamily, targetMarkers)
% ApriltagsDemo([camId][, colordepth=1][, tagSize=0.017][, tagFamily='tag16h5'][, targetMarkers])
%
% Use ApriltagsDemo to track April tag markers and visualize them in live-video.
%
% Minimalistic demo on how to capture video data and use PsychCV's builtin
% apriltag marker tracking to detect and track the rigid position and orientation
% of such fiducial markers, then visualize corresponding 3D objects on top of the
% live video of the markers via OpenGL.
%
% Note that this demo can only act as a starting point. There are too many parameters
% which need to be set for your specific use case, and we can't pass all of them via
% function parameters. Read the code, think about it, hack it accordingly.
%
% Optional parameters:
%
% camId = Id of the camera to use, [] selects default cam.
% colordepth = Video capture mode: 1 = Mono/Gray, 3 = RGB color, 4 = RGBA color.
% tagSize = Size of April tag (edge length) in meters. Needed for 3D pose estimation.
% tagFamily = April tag family to use.
% targetMarkers = List of marker id's to track and return. Default is [] for "all markers".
%
%
% Press any key to finish the demo.
%
% History:
% 15.03.2024 mk Derived from ARToolkitDemo.
% Setup defaults and normalized color range:
PsychDefaultSetup(2);
if ~IsLinux
fprintf('Sorry, this demo is currently only supported on Linux.\n');
return;
end
if nargin < 1 || isempty(camId)
% Use default camera, unless if running on machine "touchy", then use cam 2:
camId = [];
computer = Screen('Computer');
if strcmp(computer.machineName, 'touchy')
camId = 2;
end
end
% Use 1 channel mono/grayscale video for highest efficiency by default:
if nargin < 2 || isempty(colordepth)
colordepth = 1;
end
if nargin < 3 || isempty(tagSize)
tagSize = 0.017; % April tag size in meters. Default to 1.7 cm.
end
% Need to have some default tagFamily:
if nargin < 4 || isempty(tagFamily)
tagFamily = 'tag16h5';
end
% Try to detect all markers by default:
if nargin < 5 || isempty(targetMarkers)
targetMarkers = [];
end
try
% No sync tests for this simple demo:
olsync = Screen('Preference', 'SkipSyncTests', 2);
% Setup OpenGL for rendering:
InitializeMatlabOpenGL;
% Open onscreen window for diplay:
screenid = max(Screen('Screens'));
PsychImaging('PrepareConfiguration');
% Flip image horizontally to avoid confusing the viewer due to mirror view:
% PsychImaging('AddTask', 'AllViews', 'FlipHorizontal');
[win, winrect] = PsychImaging('OpenWindow', screenid, 0);
% Open videocapture device camId, at default resolution, return 1 channel grayscale images:
grabber = Screen('OpenVideoCapture', win, camId, [], colordepth);
% Start video capture at maximum possible framerate:
Screen('StartVideoCapture', grabber, realmax, 1);
% Get a single texture that we can query for the image format:
vidtex = Screen('GetCapturedImage', win, grabber);
[w, h] = Screen('WindowSize', vidtex)
channels = Screen('PixelSize', vidtex) / 8
% Compute scaling factor for upscaling all content to fill window:
[winw, winh] = Screen('WindowSize', win);
winscale = [winw / w, winh / h];
% Debug level for tracker:
PsychCV('Verbosity', 3);
% Initialize Apriltags:
% Camera intrinsics. Should come from calibration, but here we just make stuff up:
cx = w / 2;
cy = h / 2;
fx = w * 0.555;
fy = w * 0.555;
camIntrinsics = [ cx, cy, fx, fy ];
% View frustum for 3D rendering near and far distance, 0.01 - 5 meters:
zmin = 0.01;
zmax = 5;
% Setup parameters for 3D marker 6-DoF pose reconstruction, and retrieve the
% GL_PROJECTION_MATRIX to use for rendering tags superimposed onto the video:
projectionMatrix = PsychCV('April3DSettings', camIntrinsics, tagSize, zmin, zmax);
% Assign April tag family, input image size, and properties for raw video input
% images, get the imgbuffer memory buffer handle where the Psychtoolbox video
% capture engine should store camera images as input for marker detection:
imgbuffer = PsychCV('AprilInitialize', tagFamily, w, h, channels);
% Request multi-threaded operation with 8 threads, instead of the default 1 thread single-threaded:
PsychCV('AprilSettings', 8);
% Tunable parameters for 2D tag marker detection:
% [nrThreads, imageDecimation, quadSigma, refineEdges, decodeSharpening] = PsychCV('AprilSettings', 1, 2, 0.5, 1, 1)
% Some markers to visualize:
marker = [9, 11, 23];
% Size of rendered objects in meters:
sizeM = 0.01;
% Setup OpenGL 3D rendering for our simple test:
Screen('BeginOpenGL', win);
% Set projection matrix to the one provided by apriltags:
glMatrixMode(GL.PROJECTION);
glLoadMatrixd(projectionMatrix);
% Init modelview matrix to identity: Will be overwritten on each
% object draw operation by the modelview matrix from apriltag:
glMatrixMode(GL.MODELVIEW);
glLoadIdentity;
% Position the light source:
glEnable(GL.LIGHTING);
glEnable(GL.LIGHT0);
glLightfv(GL.LIGHT0, GL.POSITION, [ 0, 0, 1, 0 ]);
% Change the color - or better: The light reflection properties of the
% material - of the following objects to greenish. We only change ambient and
% diffuse reflection properties. The color for specular reflection is left
% to its default of "white":
glMaterialfv(GL.FRONT_AND_BACK,GL.AMBIENT, [ 0.0 0.6 0.0 1 ]);
glMaterialfv(GL.FRONT_AND_BACK,GL.DIFFUSE, [ 0.0 0.6 0.0 1 ]);
% Set proper viewport:
glViewport(0, 0, w * winscale(1), h * winscale(2));
% Define background clearcolor:
glClearColor(0,0,0,0);
% Enable depth test for depth correct 3D rendering:
glEnable(GL.DEPTH_TEST);
% OpenGL 3D setup done:
Screen('EndOpenGL', win);
% Init stats:
count = 0;
tstart = GetSecs;
KbReleaseWait;
% Run tracking and viz loop until keypress:
while ~KbCheck
% Retrieve next captured image in vidtex: The 'specialMode'= 4 flag
% requests video data to be put into the PsychCV() memory buffer
% referenced by 'imgbuffer':
vidtex = Screen('GetCapturedImage', win, grabber, 1, vidtex, 4, imgbuffer);
% Draw captured video image as backdrop:
Screen('DrawTexture', win, vidtex, [], ScaleRect(Screen('Rect', vidtex), winscale(1), winscale(2)));
% Clear depth buffer before rendering 3D objects for this frame:
Screen('BeginOpenGL', win);
glClear(GL.DEPTH_BUFFER_BIT);
Screen('EndOpenGL', win);
%fprintf('\n=================================\n');
% Detect apriltags in captured image:
detectedMarkers = PsychCV('AprilDetectMarkers', targetMarkers);
% For each candidate marker do...
j = 0;
for i = 1:length(detectedMarkers)
% i'th candidate detected reliably?
score = detectedMarkers(i).MatchQuality;
hamming = detectedMarkers(i).HammingErrorBits;
if score > 100 && hamming <= 2
% Yes. Get info about it:
j = j + 1;
id = detectedMarkers(i).Id;
% Visualize (x,y) 2D image position of marker center with a green dot:
xy = detectedMarkers(i).Center2D;
Screen('DrawDots', win, xy .* winscale', 10, [0 1 0], [], 1);
% Visualize (x,y) 2D image position of marker corners as yellow dots,
% and boundaries as red lines:
xyv = detectedMarkers(i).Corners2D;
Screen('FramePoly', win, [1 0 0], xyv' .* winscale, 2);
Screen('DrawDots', win, xyv .* winscale', 10, [1 1 0], [], 1);
% 3D pose and pose reconstruction error:
poseError = detectedMarkers(i).PoseError;
% Pose as OpenGL GL_MODELVIEW_MATRIX compatible 4x4 matrix:
OpenGLPoseMatrix = detectedMarkers(i).ModelViewMatrix;
% 3D position of the markers center in apriltag natural reference frame.
% Extract mpos as the 3 component translation vector from the pose matrix:
mpos = detectedMarkers(i).TransformMatrix(1:3, 4);
%fprintf('Id: %i Hamming %i Score %f PosError [mm]: %f\n', id, hamming, score, poseError * 1e3);
% Draw some text next to each markers center, with marker id and estimated 'mpos'
% 3D center coordinates in meters, relative to the optical center of the camera,
% with x-axis to the right, y-axis down, and z-axis facing away from the camera,
% ie. along its optical axis, ie. in the viewing direction:
statusTxt = sprintf('%i: %04f : %04f : %04f', id, mpos(1), mpos(2), mpos(3));
Screen('DrawText', win, statusTxt, xy(1) * winscale(1) + 10, xy(2) * winscale(2), [1 0 1], [], 0);
% Visualize detected marker in 3D as well, via OpenGL 3D rendering:
Screen('BeginOpenGL', win);
% Setup rigid position and orientation of marker via the OpenGLPoseMatrix:
glLoadMatrixd(OpenGLPoseMatrix);
% Visualize location and orientation of tag, by drawing some reference frame on it:
% Tag in x-y plane, z-axis comes out of tag like a surface normal vector.
% Draw little reference frame in 3D, ie. x/y/z axis as red, green, blue lines:
glDisable(GL.LIGHTING);
glBegin(GL.LINES);
glColor3d(1, 0, 0);
glVertex3d(0, 0, 0);
glVertex3d(sizeM, 0, 0);
glColor3d(0, 1, 0);
glVertex3d(0, 0, 0);
glVertex3d(0, sizeM, 0);
glColor3d(0, 0, 1);
glVertex3d(0, 0, 0);
glVertex3d(0, 0, sizeM);
glEnd;
% For the fun of it, also draw some 3D objects "glued" onto the
% tags denoted by the vector 'marker':
glEnable(GL.LIGHTING);
% Switch drawn 3D objects by markerId:
glTranslatef(0, 0, sizeM/2)
switch j
case 1,
glutSolidTeapot(sizeM);
case 2,
glutSolidCube(sizeM);
case 3,
glutSolidSphere(sizeM, 50, 50);
end
% Done with this markers 3D drawing:
Screen('EndOpenGL', win);
end
end
% Show updated image at next retrace:
Screen('Flip', win);
% Next tracking loop iteration...
count = count + 1;
end
% Stats...
telapsed = GetSecs - tstart;
fprintf('Average fps: %f\n', count / telapsed);
% Stop capture, close engine and onscreen window:
Screen('StopVideoCapture', grabber);
Screen('CloseVideoCapture', grabber);
sca;
% Shutdown tracker:
PsychCV('AprilShutdown');
catch
sca;
PsychCV('AprilShutdown');
end
% Restore sync settings:
Screen('Preference', 'SkipSyncTests', olsync);
|