def __init__(self, img_size):
self.img_size = img_size
self.nintr = 9
self.unknowns = None # number of unknowns in our equation system
# initial K matrix
# with aspect ratio of 1 and pp at center. Focal length is empirical.
self.Kin = cv2.getDefaultNewCameraMatrix(np.diag([1000, 1000, 1]),
img_size, True)
self.K = self.Kin.copy()
self.cdist = None
self.flags = cv2.CALIB_USE_LU
# calibration data
self.keyframes = []
self.reperr = float("NaN")
self.PCov = np.zeros((self.nintr, self.nintr)) # parameter covariance
self.pose_var = np.zeros(6)
self.disp_idx = None # index of dispersion
import numpy as np
import cv2 as cv
# aruco
adict = cv.aruco.Dictionary_get(cv.aruco.DICT_4X4_50)
cv.imshow("marker", cv.aruco.drawMarker(adict, 0, 400))
# random calibration data. your mileage may vary.
imsize = (800, 600)
K = cv.getDefaultNewCameraMatrix(np.diag([800, 800, 1]), imsize, True)
# AR scene
cv.ovis.addResourceLocation("packs/Sinbad.zip") # shipped with Ogre
win = cv.ovis.createWindow("arucoAR", imsize, flags=0)
win.setCameraIntrinsics(K, imsize)
win.createEntity("figure", "Sinbad.mesh", (0, 0, 5), (1.57, 0, 0))
win.createLightEntity("sun", (0, 0, 100))
# video capture
cap = cv.VideoCapture(0)
cap.set(cv.CAP_PROP_FRAME_WIDTH, imsize[0])
cap.set(cv.CAP_PROP_FRAME_HEIGHT, imsize[1])
while cv.ovis.waitKey(1) != 27:
img = cap.read()[1]
win.setBackground(img)
corners, ids = cv.aruco.detectMarkers(img, adict)[:2]
cv.waitKey(1)
import numpy as np
import cv2 as cv
# aruco
adict = cv.aruco.Dictionary_get(cv.aruco.DICT_4X4_50)
cv.imshow("marker", cv.aruco.drawMarker(adict, 0, 400))
# random calibration data. your mileage may vary.
imsize = (800, 600)
K = cv.getDefaultNewCameraMatrix(np.diag([800, 800, 1]), imsize, True)
# AR scene
cv.ovis.addResourceLocation("packs/Sinbad.zip") # shipped with Ogre
win = cv.ovis.createWindow("arucoAR", imsize, flags=0)
win.setCameraIntrinsics(K, imsize)
win.createEntity("figure", "Sinbad.mesh", (0, 0, 5), (1.57, 0, 0))
win.createLightEntity("sun", (0, 0, 100))
# video capture
cap = cv.VideoCapture(0)
cap.set(cv.CAP_PROP_FRAME_WIDTH, imsize[0])
cap.set(cv.CAP_PROP_FRAME_HEIGHT, imsize[1])
while cv.ovis.waitKey(1) != 27:
img = cap.read()[1]
win.setBackground(img)
corners, ids = cv.aruco.detectMarkers(img, adict)[:2]
cv.waitKey(1)
def main(ctx):
## random calibration data. your mileage may vary
imsize = (800, 600)
K = cv2.getDefaultNewCameraMatrix(np.diag([800, 800, 1]), imsize, True)
## setup Ogre for AR
scn_mgr = ctx.getRoot().createSceneManager()
Ogre.RTShader.ShaderGenerator.getSingleton().addSceneManager(scn_mgr)
cam = scn_mgr.createCamera("camera")
cam.setNearClipDistance(5)
ctx.getRenderWindow().addViewport(cam)
camnode = scn_mgr.getRootSceneNode().createChildSceneNode()
# convert OpenCV to OGRE coordinate system
camnode.rotate((1, 0, 0), math.pi)
camnode.attachObject(cam)
set_camera_intrinsics(cam, K, imsize)
bgtex = create_image_background(scn_mgr)
## setup 3D scene
scn_mgr.setAmbientLight((.1, .1, .1))
scn_mgr.getRootSceneNode().createChildSceneNode().attachObject(
scn_mgr.createLight())
marker_node = scn_mgr.getRootSceneNode().createChildSceneNode()
mesh_node = marker_node.createChildSceneNode()
mesh_node.attachObject(scn_mgr.createEntity("Sinbad.mesh"))
mesh_node.rotate((1, 0, 0), math.pi / 2)
mesh_node.translate((0, 0, 5))
mesh_node.setVisible(False)
## video capture
cap = cv2.VideoCapture(0)
cap.set(cv2.CAP_PROP_FRAME_WIDTH, imsize[0])
cap.set(cv2.CAP_PROP_FRAME_HEIGHT, imsize[1])
## aruco
adict = cv2.aruco.getPredefinedDictionary(cv2.aruco.DICT_4X4_50)
cv2.imshow("marker", cv2.aruco.drawMarker(adict, 0, 400))
rvec, tvec = None, None
while cv2.waitKey(1) != 27:
cont, img = cap.read()
if not cont: break
im = Ogre.Image(Ogre.PF_BYTE_BGR, img.shape[1], img.shape[0], 1, img,
False)
if bgtex.getBuffer():
bgtex.getBuffer().blitFromMemory(im.getPixelBox())
else:
bgtex.loadImage(im)
corners, ids = cv2.aruco.detectMarkers(img, adict)[:2]
if ids is not None:
rvecs, tvecs = cv2.aruco.estimatePoseSingleMarkers(
corners, 5, K, None)[:2]
rvec, tvec = rvecs[0].ravel(), tvecs[0].ravel()
ax = Ogre.Vector3(*rvec)
ang = ax.normalise()
marker_node.setOrientation(Ogre.Quaternion(ang, ax))
marker_node.setPosition(tvec)
mesh_node.setVisible(True)
ctx.getRoot().renderOneFrame()
import cv2
import sys
cascPath = sys.argv[1]
faceCascade = cv2.CascadeClassifier(cascPath)
video_capture = cv2.VideoCapture(1)
video_capture = cv2.getDefaultNewCameraMatrix()
while True:
# Capture frame-by-frame
ret, frame = video_capture.read()
gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
faces = faceCascade.detectMultiScale(
gray,
scaleFactor=1.2,
minNeighbors=5,
minSize=(30, 30),
flags=cv2.CASCADE_SCALE_IMAGE
# cv2.cv.CV_HAAR_SCALE_IMAGE
)
# Draw a rectangle around the faces
for (x, y, w, h) in faces:
cv2.rectangle(frame, (x, y), (x+w, y+h), (0, 255, 0), 2)
# Display the resulting frame
cv2.imshow('Video', frame)