def start(self):
try:
self.logger.info("Starting Application..")
if self.auto_reload:
self.load_reconstruction_from_pickle()
else:
self.reconstruction = Reconstruction(
slide_score_api=self.slide_score_api,
slide_score_user=self.slide_score_user,
parent=self,
logger=self.logger)
self.reconstruction.set_slide_score_study_and_case_id(
slide_score_study_id=self.slide_score_study_id,
slide_score_case_id=self.slide_score_case_id)
self.reconstruction.set_macro_photo(
macro_photo_path=self.macro_photo_path)
self.reconstruction.load_coupes(
max_cnt_coupes=self.max_cnt_coupes)
except:
self.logger.error(
f'Unexpected error: {sys.exc_info()[0]} \n {traceback.format_exc()}'
)
pass
def reconstruct(self, K, imageSize, markers):
def normalizeMarkersCostFunction(parameters):
x = parameters[0]
y = parameters[1]
r2 = x*x + y*y
r4 = r2*r2
r6 = r4*r2
r_coeff = 1 + k1*r2 + k2*r4 + k3*r6
xx = K[0, 0]*x*r_coeff + K[0, 2]
yy = K[1, 1]*y*r_coeff + K[1, 2]
return xx - m.x, yy - m.y
k1, k2, k3 = 0, 0, 0 # polynomial radial distortion
#Normalize markers
normalizedMarkers = []
for m in markers:
initParams = numpy.zeros(2)
initParams[0] = (m.x - K[0, 2]) / K[0, 0]
initParams[1] = (m.y - K[1, 2]) / K[1, 1]
result = optimize.root(normalizeMarkersCostFunction, initParams, method='lm')
#print(initParams[0], initParams[1], m.x, m.y, result.x[0], result.x[1])
normalizedMarkers.append(Marker(m.frame, m.track, result.x[0], result.x[1]))
#select two keyframes
#keyframes = self.selectKeyFrames(normalizedMarkers, K)
kf1, kf2 = 0, Marker.getMaxFrame(normalizedMarkers)
#Actual reconstruction
keyframeMarkers = Marker.markersForTracksInBothFrames(markers, kf1, kf2)
if len(keyframeMarkers) < 8:
print('Error: Not Enough Keyframe Markers.')
return None, None, None
maxFrame = Marker.getMaxFrame(markers)
maxTrack = Marker.getMaxTrack(markers)
reconstruction = Reconstruction(maxFrame, maxTrack)
self.reconstructTwoFrames(keyframeMarkers, reconstruction)
#TODO: EuclideanBundle(normalized_tracks)
self.completeReconstruction(normalizedMarkers, reconstruction)
#TODO: EuclideanBundleCommonIntrinsics
self.scaleToUnity(reconstruction)
#TODO: EuclideanReprojectionError
return reconstruction.extract() # K is refined but not returned here
def main():
parser = argparse.ArgumentParser()
parser.add_argument("--lexicon",
help="Specify the path of the lexical file.",
type=str)
parser.add_argument("--document",
help="Specify the path of the input document.",
type=str)
parser.add_argument("--output",
help="Specify the path of the output document.",
type=str)
args = parser.parse_args()
if args.lexicon is None or args.document is None or args.output is None:
parser.print_help(sys.stderr)
sys.exit(1)
if not validate_input(args):
print("Please enter valid paths for the inputs.")
parser.print_help(sys.stderr)
sys.exit(1)
doc_reconstructor = Reconstruction(args.document, args.output,
args.lexicon)
doc_reconstructor.re_construct()
parser.add_argument('--num_points',
type=int,
default=2048,
help='Num of points to use')
parser.add_argument('--model_path',
type=str,
default='',
metavar='N',
help='Path to load model')
args = parser.parse_args()
return args
if __name__ == '__main__':
args = get_parser()
if args.eval == False:
if args.task == 'reconstruct':
reconstruction = Reconstruction(args)
reconstruction.run()
elif args.task == 'classify':
classification = Classification(args)
classification.run()
elif args.task == 'segment':
segmentation = Segmentation(args)
segmentation.run()
else:
inference = Inference(args)
feature_dir = inference.run()
svm = SVM(feature_dir)
svm.run()
altitude3 = 59
wahrsis3 = Imager(name3, center3, radius3, relativePosition3, calibRot3,
calibTrans3, longitude3, lattitude3, altitude3)
name4 = 'wahrsis4'
center4 = [2000, 2975]
radius4 = 1665
relativePosition4 = np.array([-2.334, 101.3731, -8.04])
calibRot4 = np.array([[0.9710936, -0.23401871, 0.04703662],
[0.234924, 0.97190314, -0.01466276],
[-0.04228367, 0.02528894, 0.99878553]])
calibTrans4 = np.array([[-0.00274625], [-0.00316865], [0.00516088]])
wahrsis4 = Imager(name4, center4, radius4, relativePosition4, calibRot4,
calibTrans4)
images3 = [
cv2.imread('wahrsis3/2015-10-29-12-58-01-wahrsis3-low.jpg'),
cv2.imread('wahrsis3/2015-10-29-12-58-01-wahrsis3-med.jpg'),
cv2.imread('wahrsis3/2015-10-29-12-58-01-wahrsis3-high.jpg')
]
images4 = [
cv2.imread('wahrsis4/2015-10-29-12-58-01-wahrsis4-low.jpg'),
cv2.imread('wahrsis4/2015-10-29-12-58-01-wahrsis4-med.jpg'),
cv2.imread('wahrsis4/2015-10-29-12-58-01-wahrsis4-high.jpg')
]
reconst = Reconstruction(wahrsis3, wahrsis4,
datetime.datetime(2015, 10, 29, 12, 58, 1))
reconst.load_images(images3, images4)
reconst.process()
reconst.writeFigure("out.png")
from reconstruction import Reconstruction
pixel_height = 192
pixel_width = 256
if __name__ == '__main__':
rec = Reconstruction(image_width=pixel_width,
image_height=pixel_height,
train_directory='test_imgs')
print("End test")