ax.axhline(0, color='grey', ls='dashed')
#ax.set_title('q{}'.format(i))
def plot_position(ax, i):
ax.plot(positions_gt[:, i], color='grey')
ax.plot(positions_test[:, i])
#ax.yaxis.set_ticks(np.arange(min_pos, max_pos, 1))
ax.set_ylim(
min(
0,
math.floor(min(min(positions_gt[:, i]), min(positions_test[:, i])))
- 1))
_, labels_gt = read_label_file(def_file, convert_to_axis=True)
_, labels_test = read_label_file(test_file, convert_to_axis=False)
positions_gt = np.array(map(lambda x: x[0:3], labels_gt))
positions_test = np.array(map(lambda x: x[0:3], labels_test))
orientations_gt = np.array(map(lambda x: x[3:], labels_gt))
orientations_test = np.array(map(lambda x: x[3:], labels_test))
if orientations_gt.shape[1] == 3:
pass
elif flip == 'auto':
for i, x in enumerate(orientations_gt):
if sum(abs(-x - orientations_test[i])) < sum(
abs(x - orientations_test[i])):
orientations_gt[i] = -x
parser.add_argument('--arrow_len',
action='store',
type=float,
required=False,
default=1)
parser.add_argument('--connect',
action='store_true',
help='''Connect consecutive camera positions with lines''')
parser.add_argument('--rings', action='store', nargs='*', required=False)
parser.add_argument('--plot_gt', action='store_true')
parser.add_argument('--plot_diff', action='store_true')
parser.add_argument('-u', '--uncertainty', action='store_true')
args = parser.parse_args()
if not args.model or args.plot_gt or args.plot_diff:
_, labels = read_label_file(args.dataset)
positions_gt = np.array([l[0:3] for l in labels])
orientations_gt = np.array([l[3:7] for l in labels])
if args.model:
input_size = 256
test_reader = ImageReader(args.dataset,
batch_size=1,
image_size=[input_size, input_size],
randomise=False)
n_images = test_reader.total_images()
positions = np.empty([0, 3])
orientations = np.empty([0, 4])
if args.uncertainty:
std_x = []
plt.plot(pos[0] + radius * np.cos(t),
pos[1] + radius * np.sin(t),
color,
marker='.',
lw=0,
ms=10)
def_file = 'results_new/david_6a/test2_localised.txt'
david = True
side = True
extrapolation = False
radius = 3.5 #*np.sin(1.2)
lim = 4.5
_, labels = read_label_file(def_file, convert_to_axis=False)
positions = np.array([l[0:3] for l in labels])
orientations = np.array([l[3:] for l in labels])
if side and david:
positions_2d = [(np.sqrt(p[0] * p[0] + p[1] * p[1]), p[2])
for p in positions]
orientations_2d = [(-1, 0) for o in orientations]
elif side:
positions_2d = [((p[0] + p[1]) / np.sqrt(2), p[2]) for p in positions]
orientations_2d = [((o[0] + o[1]) / np.sqrt(2), o[2])
for o in orientations]
else:
positions_2d = [(p[0], p[1]) for p in positions]
orientations_2d = [(o[0], o[1]) for o in orientations]
from posenet.utils import progress_bar
parser = argparse.ArgumentParser()
parser.add_argument('-m', '--model', action='store', required=True,
help='''Path to a trained Tensorflow model (.ckpt file)''')
parser.add_argument('-d', '--dataset', action='store', required=True,
help='''Path to a text file listing images and camera poses''')
parser.add_argument('-o', '--output', action='store', required=False)
args = parser.parse_args()
if os.path.isdir(args.dataset):
imgs = glob.glob('{}/*.png'.format(args.dataset))
elif args.dataset.endswith('.txt'):
imgs, _ = read_label_file(args.dataset, full_paths=True)
elif args.dataset.endswith('.png'):
imgs = [args.dataset]
from matplotlib import pyplot as plt
else:
imgs = []
n_images = len(imgs)
if n_images > 1 and not args.output:
print('--output argument required')
sys.exit(1)
if n_images == 0:
print('No images found')
sys.exit(1)
required=True,
help='''Path to a text file listing images and camera poses''')
parser.add_argument('-o', '--output', action='store', required=False)
parser.add_argument('-u', '--uncertainty', action='store_true')
args = parser.parse_args()
input_size = 224
test_reader = ImageReader(args.dataset,
batch_size=1,
image_size=[input_size, input_size],
random_crop=False,
randomise=False)
n_images = test_reader.total_images()
# Read the definition file to get file names
paths, _ = read_label_file(args.dataset)
paths = map(lambda x: os.path.basename(x), paths)
# Localise
with Localiser(input_size, args.model,
uncertainty=args.uncertainty) as localiser:
if args.output:
f = open(args.output, 'w')
f.write('\n\n\n') # Hack for now
for i in range(n_images):
images_feed, labels_feed = test_reader.next_batch()
# Make prediction
predicted = localiser.localise(images_feed)
x = [round(v, 6) for v in predicted['x']]
if args.agg:
matplotlib.use('Agg')
import matplotlib.colors as colors
import matplotlib.cm as cmx
from mpl_toolkits.mplot3d import Axes3D
import matplotlib.pyplot as plt
from posenet.core.image_reader import ImageReader, read_label_file
from posenet.core.localiser import Localiser
from posenet.utils import progress_bar
from posenet.utils import rotate_by_quaternion
output_type = 'axis' if args.axis else 'quat'
if not args.model or args.plot_gt or args.plot_diff:
_, labels = read_label_file(args.dataset,
convert_to_axis=args.convert_to_axis)
positions_gt = np.array([l[0:3] for l in labels])
orientations_gt = np.array([l[3:] for l in labels])
if args.model:
input_size = [256, 455]
crop_size = [256, 256]
test_reader = ImageReader(args.dataset,
batch_size=1,
image_size=input_size,
crop_size=crop_size,
centre_crop=True,
randomise=False)
n_images = test_reader.total_images()
positions = []