def image_cb(self, msg):
img_orig = self.bridge.imgmsg_to_cv2(msg, desired_encoding='rgb8')
img = np.asarray(img_orig) / 255
H, W = img.shape[:2]
h_step = self.patch_height * self.subsample
w_step = self.patch_width * self.subsample
h_idxs = np.arange(0, H, h_step)[:-1]
w_idxs = np.arange(0, W, h_step)[:-1]
results = []
for y in h_idxs:
for x in w_idxs:
sub_img = img[y:y+h_step:self.subsample, x:x+w_step:self.subsample, :][np.newaxis, ...]
self.interpreter.set_tensor(self.input_details[0]['index'], sub_img.astype(np.float32))
self.interpreter.invoke()
cls = self.interpreter.get_tensor(self.output_details[0]['index'])
cls = [np.argmax(cls), np.max(cls)]
results.append(cls)
labels = np.asarray([p[0] for p in results])
probs = np.asarray([p[1] for p in results])
cls_msg = ClassificationResult(
header=msg.header,
classifier=self.classifier_name,
target_names=self.label_names,
labels=labels,
label_names=[self.label_names[l] for l in labels],
label_proba=probs)
self.pub_classification.publish(cls_msg)
if self.visualize:
fig = plt.figure(tight_layout={'pad': 0})
ax = plt.Axes(fig, [0., 0., 1., 1.])
ax.axis('off')
fig.add_axes(ax)
clr = np.asarray([[1,0,0], [0,1,0], [0,0,1], [0,0,0]])
idx = 0
ax = vis_image(img_orig.transpose((2, 0, 1)), ax=ax)
for y in h_idxs:
for x in w_idxs:
ax.add_patch(plt.Rectangle(
(x,y),
self.patch_width * self.subsample,
self.patch_height * self.subsample,
fill=True,
facecolor=clr[labels[idx]],
linewidth=0,
alpha=0.3))
idx += 1
fig.canvas.draw()
w, h = fig.canvas.get_width_height()
vis_img = np.fromstring(
fig.canvas.tostring_rgb(), dtype=np.uint8)
vis_img.shape = (h, w, 3)
fig.clf()
plt.close()
vis_msg = self.bridge.cv2_to_imgmsg(vis_img, 'rgb8')
# BUG: https://answers.ros.org/question/316362/sensor_msgsimage-generates-float-instead-of-int-with-python3/ # NOQA
vis_msg.step = int(vis_msg.step)
vis_msg.header = msg.header
self.pub_image.publish(vis_msg)
def vis_bbox(img, bbox, label=None, score=None, label_names=None,
instance_colors=None, alpha=1., linewidth=3.,
sort_by_score=True, ax=None):
from matplotlib import pyplot as plt
if label is not None and not len(bbox) == len(label):
raise ValueError('The length of label must be same as that of bbox')
if score is not None and not len(bbox) == len(score):
raise ValueError('The length of score must be same as that of bbox')
if sort_by_score and score is not None:
order = np.argsort(score)
bbox = bbox[order]
score = score[order]
if label is not None:
label = label[order]
if instance_colors is not None:
instance_colors = np.array(instance_colors)[order]
# Returns newly instantiated matplotlib.axes.Axes object if ax is None
ax = vis_image(img, ax=ax)
# If there is no bounding box to display, visualize the image and exit.
if len(bbox) == 0:
return ax
if instance_colors is None:
# Red
instance_colors = np.zeros((len(bbox), 3), dtype=np.float32)
instance_colors[:, 0] = 255
instance_colors = np.array(instance_colors)
for i, bb in enumerate(bbox):
xy = (bb[1], bb[0])
height = bb[2] - bb[0]
width = bb[3] - bb[1]
color = instance_colors[i % len(instance_colors)] / 255
ax.add_patch(plt.Rectangle(
xy, width, height, fill=False,
edgecolor=color, linewidth=linewidth, alpha=alpha))
caption = []
if label is not None and label_names is not None:
lb = label[i]
if not (0 <= lb < len(label_names)):
raise ValueError('No corresponding name is given')
caption.append(label_names[lb])
if score is not None:
sc = score[i]
caption.append(u"{:.2f}".format(sc))
if len(caption) > 0:
ax.text(bb[1], bb[0],
u": ".join(caption),
style='italic',
bbox={'facecolor': 'white', 'alpha': 0.7, },
fontsize=5)
return ax
def vis_point(img, point, visible=None, ax=None):
"""Visualize points in an image.
Example:
>>> import chainercv
>>> import matplotlib.pyplot as plt
>>> dataset = chainercv.datasets.CUBKeypointDataset()
>>> img, point, visible = dataset[0]
>>> chainercv.visualizations.vis_point(img, point, visible)
>>> plt.show()
Args:
img (~numpy.ndarray): See the table below.
If this is :obj:`None`, no image is displayed.
point (~numpy.ndarray or list of arrays): See the table below.
visible (~numpy.ndarray or list of arrays): See the table below.
ax (matplotlib.axes.Axes): If provided, plot on this axis.
.. csv-table::
:header: name, shape, dtype, format
:obj:`img`, ":math:`(3, H, W)`", :obj:`float32`, \
"RGB, :math:`[0, 255]`"
:obj:`point`, ":math:`(R, K, 2)` or :math:`[(K, 2)]`", \
:obj:`float32`, ":math:`(y, x)`"
:obj:`visible`, ":math:`(R, K)` or :math:`[(K,)]`", :obj:`bool`, --
Returns:
~matploblib.axes.Axes:
Returns the Axes object with the plot for further tweaking.
"""
import matplotlib.pyplot as plt
# Returns newly instantiated matplotlib.axes.Axes object if ax is None
ax = vis_image(img, ax=ax)
_, H, W = img.shape
n_inst = len(point)
cm = plt.get_cmap('gist_rainbow')
for i in range(n_inst):
pnt = point[i]
n_point = len(pnt)
if visible is not None:
vsbl = visible[i]
else:
vsbl = np.ones((n_point, ), dtype=np.bool)
colors = [cm(k / n_point) for k in six.moves.range(n_point)]
for k in range(n_point):
if vsbl[k]:
ax.scatter(pnt[k][1], pnt[k][0], c=colors[k], s=100)
ax.set_xlim(left=0, right=W)
ax.set_ylim(bottom=H - 1, top=0)
return ax
def vis_point(img, point, ax=None):
# Returns newly instantiated matplotlib.axes.Axes object if ax is None
ax = vis_image(img, ax=ax)
for i in range(len(point)):
for j in range(len(point[i])):
ax.scatter(point[i][j][0], point[i][j][1])
return ax
def vis_keypoint(img, keypoint, kp_mask=None, ax=None):
"""Visualize keypoints in an image.
Example:
>>> import chainercv
>>> import matplotlib.pyplot as plot
>>> dataset = chainercv.datasets.CUBKeypointDataset()
>>> img, keypoint, kp_mask = dataset[0]
>>> chainercv.visualizations.vis_keypoint(img, keypoint, kp_mask)
>>> plot.show()
Args:
img (~numpy.ndarray): An image of shape :math:`(3, height, width)`.
This is in RGB format and the range of its value is
:math:`[0, 255]`. This should be visualizable using
:obj:`matplotlib.pyplot.imshow(img)`
keypoint (~numpy.ndarray): An array with keypoint pairs whose shape is
:math:`(K, 2)`, where :math:`K` is
the number of keypoints in the array.
The second axis corresponds to :math:`y` and :math:`x` coordinates
of the keypoint.
kp_mask (~numpy.ndarray, optional): A boolean array whose shape is
:math:`(K,)`. If :math:`i` th index is :obj:`True`, the
:math:`i` th keypoint is not displayed. If not specified,
all keypoints in :obj:`keypoint` will be displayed.
ax (matplotlib.axes.Axes, optional): If provided, plot on this axis.
Returns:
~matploblib.axes.Axes:
Returns the Axes object with the plot for further tweaking.
"""
import matplotlib.pyplot as plot
# Returns newly instantiated matplotlib.axes.Axes object if ax is None
ax = vis_image(img, ax=ax)
_, H, W = img.shape
n_kp = len(keypoint)
if kp_mask is None:
kp_mask = np.ones((n_kp,), dtype=np.bool)
cm = plot.get_cmap('gist_rainbow')
colors = [cm(1. * i / n_kp) for i in six.moves.range(n_kp)]
for i in range(n_kp):
if kp_mask[i]:
ax.scatter(keypoint[i][1], keypoint[i][0], c=colors[i], s=100)
ax.set_xlim(left=0, right=W)
ax.set_ylim(bottom=H - 1, top=0)
return ax
def vis_point(img, point, mask=None, ax=None):
"""Visualize points in an image.
Example:
>>> import chainercv
>>> import matplotlib.pyplot as plot
>>> dataset = chainercv.datasets.CUBKeypointDataset()
>>> img, point, mask = dataset[0]
>>> chainercv.visualizations.vis_point(img, point, mask)
>>> plot.show()
Args:
img (~numpy.ndarray): An image of shape :math:`(3, height, width)`.
This is in RGB format and the range of its value is
:math:`[0, 255]`. This should be visualizable using
:obj:`matplotlib.pyplot.imshow(img)`
point (~numpy.ndarray): An array of point coordinates whose shape is
:math:`(P, 2)`, where :math:`P` is
the number of points.
The second axis corresponds to :math:`y` and :math:`x` coordinates
of the points.
mask (~numpy.ndarray): A boolean array whose shape is
:math:`(P,)`. If :math:`i` th element is :obj:`True`, the
:math:`i` th point is not displayed. If not specified,
all points in :obj:`point` will be displayed.
ax (matplotlib.axes.Axes): If provided, plot on this axis.
Returns:
~matploblib.axes.Axes:
Returns the Axes object with the plot for further tweaking.
"""
import matplotlib.pyplot as plot
# Returns newly instantiated matplotlib.axes.Axes object if ax is None
ax = vis_image(img, ax=ax)
_, H, W = img.shape
n_point = len(point)
if mask is None:
mask = np.ones((n_point,), dtype=np.bool)
cm = plot.get_cmap('gist_rainbow')
colors = [cm(i / n_point) for i in six.moves.range(n_point)]
for i in range(n_point):
if mask[i]:
ax.scatter(point[i][1], point[i][0], c=colors[i], s=100)
ax.set_xlim(left=0, right=W)
ax.set_ylim(bottom=H - 1, top=0)
return ax
def vis_tracking_bbox(img,
bbox,
inst_id,
label=None,
score=None,
label_names=None,
alpha=1.0,
ax=None):
from matplotlib import pyplot as plot
ax = vis_image(img, ax=ax)
assert len(bbox) == len(inst_id)
if len(bbox) == 0:
return ax
for i, (bb, inst_i) in enumerate(zip(bbox, inst_id)):
bb = np.round(bb).astype(np.int32)
y_min, x_min, y_max, x_max = bb
color = np.array(_default_cmap(inst_i + 1)) / 255.
ax.add_patch(
plot.Rectangle((x_min, y_min),
x_max - x_min,
y_max - y_min,
fill=False,
edgecolor=color,
linewidth=3))
caption = []
caption.append('{}'.format(inst_i))
if label is not None and label_names is not None:
lb = label[i]
if not (0 <= lb < len(label_names)):
raise ValueError('No corresponding name is given')
caption.append(label_names[lb])
if score is not None:
sc = score[i]
caption.append('{:.2f}'.format(sc))
ax.text((x_max + x_min) / 2,
y_min,
': '.join(caption),
style='italic',
bbox={
'facecolor': color,
'alpha': alpha
},
fontsize=8,
color='white')
def vis_bbox(img,
bbox,
label=None,
score=None,
label_names=None,
instance_colors=None,
alpha=1.,
linewidth=3.,
sort_by_score=True,
ax=None):
"""Visualize bounding boxes inside image.
Example:
>>> from chainercv.datasets import VOCBboxDataset
>>> from chainercv.datasets import voc_bbox_label_names
>>> from chainercv.visualizations import vis_bbox
>>> import matplotlib.pyplot as plt
>>> dataset = VOCBboxDataset()
>>> img, bbox, label = dataset[60]
>>> vis_bbox(img, bbox, label,
... label_names=voc_bbox_label_names)
>>> plt.show()
This example visualizes by displaying the same colors for bounding
boxes assigned to the same labels.
>>> from chainercv.datasets import VOCBboxDataset
>>> from chainercv.datasets import voc_bbox_label_names
>>> from chainercv.visualizations import vis_bbox
>>> from chainercv.visualizations.colormap import voc_colormap
>>> import matplotlib.pyplot as plt
>>> dataset = VOCBboxDataset()
>>> img, bbox, label = dataset[61]
>>> colors = voc_colormap(label + 1)
>>> vis_bbox(img, bbox, label,
... label_names=voc_bbox_label_names,
... instance_colors=colors)
>>> plt.show()
Args:
img (~numpy.ndarray): See the table below. If this is :obj:`None`,
no image is displayed.
bbox (~numpy.ndarray): See the table below.
label (~numpy.ndarray): See the table below. This is optional.
score (~numpy.ndarray): See the table below. This is optional.
label_names (iterable of strings): Name of labels ordered according
to label ids. If this is :obj:`None`, labels will be skipped.
instance_colors (iterable of tuples): List of colors.
Each color is RGB format and the range of its values is
:math:`[0, 255]`. The :obj:`i`-th element is the color used
to visualize the :obj:`i`-th instance.
If :obj:`instance_colors` is :obj:`None`, the red is used for
all boxes.
alpha (float): The value which determines transparency of the
bounding boxes. The range of this value is :math:`[0, 1]`.
linewidth (float): The thickness of the edges of the bounding boxes.
sort_by_score (bool): When :obj:`True`, instances with high scores
are always visualized in front of instances with low scores.
ax (matplotlib.axes.Axis): The visualization is displayed on this
axis. If this is :obj:`None` (default), a new axis is created.
.. csv-table::
:header: name, shape, dtype, format
:obj:`img`, ":math:`(3, H, W)`", :obj:`float32`, \
"RGB, :math:`[0, 255]`"
:obj:`bbox`, ":math:`(R, 4)`", :obj:`float32`, \
":math:`(y_{min}, x_{min}, y_{max}, x_{max})`"
:obj:`label`, ":math:`(R,)`", :obj:`int32`, \
":math:`[0, \#fg\_class - 1]`"
:obj:`score`, ":math:`(R,)`", :obj:`float32`, --
Returns:
~matploblib.axes.Axes:
Returns the Axes object with the plot for further tweaking.
"""
from matplotlib import pyplot as plt
if label is not None and not len(bbox) == len(label):
raise ValueError('The length of label must be same as that of bbox')
if score is not None and not len(bbox) == len(score):
raise ValueError('The length of score must be same as that of bbox')
if sort_by_score and score is not None:
order = np.argsort(score)
bbox = bbox[order]
score = score[order]
if label is not None:
label = label[order]
if instance_colors is not None:
instance_colors = np.array(instance_colors)[order]
# Returns newly instantiated matplotlib.axes.Axes object if ax is None
ax = vis_image(img, ax=ax)
# If there is no bounding box to display, visualize the image and exit.
if len(bbox) == 0:
return ax
if instance_colors is None:
# Red
instance_colors = np.zeros((len(bbox), 3), dtype=np.float32)
instance_colors[:, 0] = 255
instance_colors = np.array(instance_colors)
for i, bb in enumerate(bbox):
xy = (bb[1], bb[0])
height = bb[2] - bb[0]
width = bb[3] - bb[1]
color = instance_colors[i % len(instance_colors)] / 255
ax.add_patch(
plt.Rectangle(xy,
width,
height,
fill=False,
edgecolor=color,
linewidth=linewidth,
alpha=alpha))
caption = []
if label is not None and label_names is not None:
lb = label[i]
if not (0 <= lb < len(label_names)):
raise ValueError('No corresponding name is given')
caption.append(label_names[lb])
if score is not None:
sc = score[i]
caption.append('{:.2f}'.format(sc))
if len(caption) > 0:
ax.text(bb[1],
bb[0],
': '.join(caption),
style='italic',
bbox={
'facecolor': 'white',
'alpha': 0.7,
'pad': 10
})
return ax
def vis_bbox(img,
bbox,
label=None,
score=None,
label_names=None,
instance_colors=None,
alpha=0.9,
linewidth=3.0,
ax=None):
"""Visualize bounding boxes inside image.
Example:
>>> from chainercv.datasets import VOCBboxDataset
>>> from chainercv.datasets import voc_bbox_label_names
>>> from light_head_rcnn.visualizations import vis_bbox
>>> import matplotlib.pyplot as plt
>>> dataset = VOCBboxDataset()
>>> img, bbox, label = dataset[60]
>>> vis_bbox(img, bbox, label,
... label_names=voc_bbox_label_names)
>>> plt.show()
This example visualizes by displaying the same colors for bounding
boxes assigned to the same labels.
>>> from chainercv.datasets import VOCBboxDataset
>>> from chainercv.datasets import voc_bbox_label_names
>>> from chainercv.visualizations.colormap import voc_colormap
>>> from light_head_rcnn.visualizations import vis_bbox
>>> import matplotlib.pyplot as plt
>>> dataset = VOCBboxDataset()
>>> img, bbox, label = dataset[61]
>>> colors = voc_colormap(label + 1)
>>> vis_bbox(img, bbox, label,
... label_names=voc_bbox_label_names,
... instance_colors=colors)
>>> plt.show()
Args:
img (~numpy.ndarray): An array of shape :math:`(3, height, width)`.
This is in RGB format and the range of its value is
:math:`[0, 255]`. If this is :obj:`None`, no image is displayed.
bbox (~numpy.ndarray): An array of shape :math:`(R, 4)`, where
:math:`R` is the number of bounding boxes in the image.
Each element is organized
by :math:`(y_{min}, x_{min}, y_{max}, x_{max})` in the second axis.
label (~numpy.ndarray): An integer array of shape :math:`(R,)`.
The values correspond to id for label names stored in
:obj:`label_names`. This is optional.
score (~numpy.ndarray): A float array of shape :math:`(R,)`.
Each value indicates how confident the prediction is.
This is optional.
label_names (iterable of strings): Name of labels ordered according
to label ids. If this is :obj:`None`, labels will be skipped.
instance_colors (iterable of tuples): List of colors.
Each color is RGB format and the range of its values is
:math:`[0, 255]`. The :obj:`i`-th element is the color used
to visualize the :obj:`i`-th instance.
If :obj:`instance_colors` is :obj:`None`, the red is used for
all boxes.
alpha (float): The value which determines transparency of the
bounding boxes. The range of this value is :math:`[0, 1]`.
linewidth (float): The thickness of the edges of the bounding boxes.
ax (matplotlib.axes.Axis): The visualization is displayed on this
axis. If this is :obj:`None` (default), a new axis is created.
Returns:
~matploblib.axes.Axes:
Returns the Axes object with the plot for further tweaking.
"""
from matplotlib import pyplot as plt
if label is not None and not len(bbox) == len(label):
raise ValueError('The length of label must be same as that of bbox')
if score is not None and not len(bbox) == len(score):
raise ValueError('The length of score must be same as that of bbox')
# Returns newly instantiated matplotlib.axes.Axes object if ax is None
ax = vis_image(img, ax=ax)
# If there is no bounding box to display, visualize the image and exit.
if len(bbox) == 0:
return ax
n_inst = len(bbox)
if instance_colors is None:
instance_colors = voc_colormap(list(range(1, n_inst + 1)))
instance_colors = np.array(instance_colors)
for i, bb in enumerate(bbox):
bb = np.round(bb).astype(np.int32)
y_min, x_min, y_max, x_max = bb
color = instance_colors[i % len(instance_colors)] / 255
ax.add_patch(
plt.Rectangle((x_min, y_min),
x_max - x_min,
y_max - y_min,
fill=False,
edgecolor=color,
linewidth=linewidth,
alpha=alpha))
caption = []
if label is not None and label_names is not None:
lb = label[i]
if not (0 <= lb < len(label_names)):
raise ValueError('No corresponding name is given')
caption.append(label_names[lb])
if score is not None:
sc = score[i]
caption.append('{:.2f}'.format(sc))
if len(caption) > 0:
ax.text(x_min,
y_min,
': '.join(caption),
style='italic',
bbox={
'facecolor': color,
'alpha': alpha
},
fontsize=8,
color='white')
return ax
def vis_bbox(img, bbox, label=None, score=None, label_names=None, ax=None):
"""Visualize bounding boxes inside image.
Example:
>>> from chainercv.datasets import VOCDetectionDataset
>>> from chainercv.datasets import voc_bbox_label_names
>>> from chainercv.visualizations import vis_bbox
>>> import matplotlib.pyplot as plot
>>> dataset = VOCDetectionDataset()
>>> img, bbox, label = dataset[60]
>>> vis_bbox(img, bbox, label,
... label_names=voc_bbox_label_names)
>>> plot.show()
Args:
img (~numpy.ndarray): An array of shape :math:`(3, height, width)`.
This is in RGB format and the range of its value is
:math:`[0, 255]`.
bbox (~numpy.ndarray): An array of shape :math:`(R, 4)`, where
:math:`R` is the number of bounding boxes in the image.
Each element is organized
by :math:`(y_{min}, x_{min}, y_{max}, x_{max})` in the second axis.
label (~numpy.ndarray): An integer array of shape :math:`(R,)`.
The values correspond to id for label names stored in
:obj:`label_names`. This is optional.
score (~numpy.ndarray): A float array of shape :math:`(R,)`.
Each value indicates how confident the prediction is.
This is optional.
label_names (iterable of strings): Name of labels ordered according
to label ids. If this is :obj:`None`, labels will be skipped.
ax (matplotlib.axes.Axis): The visualization is displayed on this
axis. If this is :obj:`None` (default), a new axis is created.
Returns:
~matploblib.axes.Axes:
Returns the Axes object with the plot for further tweaking.
"""
from matplotlib import pyplot as plot
if label is not None and not len(bbox) == len(label):
raise ValueError('The length of label must be same as that of bbox')
if score is not None and not len(bbox) == len(score):
raise ValueError('The length of score must be same as that of bbox')
# Returns newly instantiated matplotlib.axes.Axes object if ax is None
ax = vis_image(img, ax=ax)
# If there is no bounding box to display, visualize the image and exit.
if len(bbox) == 0:
return ax
for i, bb in enumerate(bbox):
xy = (bb[1], bb[0])
height = bb[2] - bb[0]
width = bb[3] - bb[1]
ax.add_patch(
plot.Rectangle(xy,
width,
height,
fill=False,
edgecolor='red',
linewidth=3))
caption = []
if label is not None and label_names is not None:
lb = label[i]
if not (0 <= lb < len(label_names)):
raise ValueError('No corresponding name is given')
caption.append(label_names[lb])
if score is not None:
sc = score[i]
caption.append('{:.2f}'.format(sc))
if len(caption) > 0:
ax.text(bb[1],
bb[0],
': '.join(caption),
style='italic',
bbox={
'facecolor': 'white',
'alpha': 0.7,
'pad': 10
})
return ax
def vis_bbox(img,
bbox,
roi,
label=None,
score=None,
mask=None,
label_names=None,
ax=None,
contour=False,
labeldisplay=True):
"""Visualize bounding boxes inside image.
Example:
>>> from chainercv.datasets import VOCDetectionDataset
>>> from chainercv.datasets import voc_bbox_label_names
>>> from chainercv.visualizations import vis_bbox
>>> import matplotlib.pyplot as plot
>>> dataset = VOCDetectionDataset()
>>> img, bbox, label = dataset[60]
>>> vis_bbox(img, bbox, label,
... label_names=voc_bbox_label_names)
>>> plot.show()
Args:
img (~numpy.ndarray): An array of shape :math:`(3, height, width)`.
This is in RGB format and the range of its value is
:math:`[0, 255]`.
bbox (~numpy.ndarray): An array of shape :math:`(R, 4)`, where
:math:`R` is the number of bounding boxes in the image.
Each element is organized
by :obj:`(y_min, x_min, y_max, x_max)` in the second axis.
label (~numpy.ndarray): An integer array of shape :math:`(R,)`.
The values correspond to id for label names stored in
:obj:`label_names`. This is optional.
score (~numpy.ndarray): A float array of shape :math:`(R,)`.
Each value indicates how confident the prediction is.
This is optional.
label_names (iterable of strings): Name of labels ordered according
to label ids. If this is :obj:`None`, labels will be skipped.
ax (matplotlib.axes.Axis): The visualization is displayed on this
axis. If this is :obj:`None` (default), a new axis is created.
Returns:
~matploblib.axes.Axes:
Returns the Axes object with the plot for further tweaking.
"""
from matplotlib import pyplot as plot
if label is not None and not len(bbox) == len(label):
raise ValueError('The length of label must be same as that of bbox')
if score is not None and not len(bbox) == len(score):
raise ValueError('The length of score must be same as that of bbox')
# Returns newly instantiated matplotlib.axes.Axes object if ax is None
ax = vis_image(img, ax=ax)
# If there is no bounding box to display, visualize the image and exit.
if len(bbox) == 0:
return ax
COLOR = [(1, 1, 0), (1, 0, 1), (0, 1, 1), (0, 0, 1), (0, 1, 0), (1, 0, 0),
(0.1, 1, 0.2)]
for i, (bb, r) in enumerate(zip(bbox, roi)):
#print(label[i])
#if label[i] >1:
# continue
xy = (bb[1], bb[0])
height = int(bb[2]) - int(bb[0])
width = int(bb[3]) - int(bb[1])
ax.add_patch(
plot.Rectangle(xy,
width,
height,
fill=False,
edgecolor='red',
linewidth=1))
if mask is not None:
M = mask[i]
padded_mask = np.zeros((img.shape[2], img.shape[1]),
dtype=np.uint8)
resized_mask = cv2.resize(mask[i].T * 255, (height, width))
padded_mask[int(bb[1]):int(bb[3]),
int(bb[0]):int(bb[2])] = resized_mask
Mcontours = find_contours(padded_mask / 255, 0.3)
for verts in Mcontours:
p = Polygon(verts, facecolor="none", edgecolor=[1, 1, 1])
#print(M)
caption = list()
for my in range(14):
for mx in range(14):
mxy = (r[1] + (r[3] - r[1]) / 14 * mx,
r[0] + (r[2] - r[0]) / 14 * my)
Mcolor = np.clip((M[my, mx]) * 1, 0, 0.5)
#print(Mcolor)
ax.add_patch(
plot.Rectangle(mxy,
int((r[3] - r[1]) / 14) + 1,
int((r[2] - r[0]) / 14) + 1,
fill=True,
linewidth=0,
facecolor=COLOR[i % len(COLOR)],
alpha=Mcolor))
if contour:
ax.add_patch(p)
if label is not None and label_names is not None:
lb = label[i]
print(lb)
if not (0 <= lb < len(label_names)):
raise ValueError('No corresponding name is given')
caption.append(label_names[lb])
if score is not None:
sc = score[i]
caption.append('{:.2f}'.format(sc))
if len(caption) > 0 and labeldisplay:
ax.text(bb[1],
bb[0],
': '.join(caption),
style='italic',
fontsize=8,
bbox={
'facecolor': 'white',
'alpha': 0.7,
'pad': 10
})
return ax
def vis_keypoint_coco(img,
point,
visible=None,
point_score=None,
thresh=2,
markersize=3,
linewidth=1,
ax=None):
"""Visualize keypoints organized as in COCO.
Example:
>>> from chainercv.datasets import COCOKeypointDataset
>>> from chainercv.visualizations import vis_keypoint_coco
>>> import matplotlib.pyplot as plt
>>> data = COCOKeypointDataset(split='val')
>>> img, point, visible = data[10][:3]
>>> vis_keypoint_coco(img, point, visible)
>>> plt.show()
Args:
img (~numpy.ndarray): See the table below.
If this is :obj:`None`, no image is displayed.
point (~numpy.ndarray): See the table below.
visible (~numpy.ndarray): See the table below. If this is
:obj:`None`, all points are assumed to be visible.
point_score (~numpy.ndarray): See the table below. If this
is :obj:`None`, the confidence of all points is infinitely
large.
thresh (float): Points with confidence below :obj:`thresh` are
not visualized.
markersize (float): The size of vertices.
linewidth (float): The thickness of edges.
ax (matplotlib.axes.Axis): The visualization is displayed on this
axis. If this is :obj:`None` (default), a new axis is created.
.. csv-table::
:header: name, shape, dtype, format
:obj:`img`, ":math:`(3, H, W)`", :obj:`float32`, \
"RGB, :math:`[0, 255]`"
:obj:`point`, ":math:`(R, K, 2)`", :obj:`float32`, \
":math:`(y, x)`"
:obj:`visible`, ":math:`(R, K)`", :obj:`bool`, \
"true when a keypoint is visible."
:obj:`point_score`, ":math:`(R, K)`", :obj:`float32`, --
Returns:
~matploblib.axes.Axes:
Returns the Axes object with the plot for further tweaking.
"""
from matplotlib import pyplot as plt
# Returns newly instantiated matplotlib.axes.Axes object if ax is None
ax = vis_image(img, ax=ax)
cmap = plt.get_cmap('rainbow')
colors = [cmap(i) for i in np.linspace(0, 1, len(coco_point_skeleton) + 2)]
if point_score is None:
point_score = np.inf * np.ones(point.shape[:2], dtype=np.float32)
if point_score.shape != point.shape[:2]:
raise ValueError('Mismatch in the number of instances or joints.')
if point.shape[1:] != (len(coco_keypoint_names[human_id]), 2):
raise ValueError('point has invisible shape')
if visible is not None:
if visible.dtype != np.bool:
raise ValueError('The dtype of `visible` should be np.bool')
if visible.shape != point.shape[:2]:
raise ValueError('Mismatch in the number of instances or joints.')
for i, vld in enumerate(visible):
point_score[i, np.logical_not(vld)] = -np.inf
for pnt, pnt_sc in zip(point, point_score):
for l in range(len(coco_point_skeleton)):
i0 = coco_point_skeleton[l][0]
i1 = coco_point_skeleton[l][1]
s0 = pnt_sc[i0]
y0 = pnt[i0, 0]
x0 = pnt[i0, 1]
s1 = pnt_sc[i1]
y1 = pnt[i1, 0]
x1 = pnt[i1, 1]
if s0 > thresh and s1 > thresh:
line = ax.plot([x0, x1], [y0, y1])
plt.setp(line, color=colors[l], linewidth=linewidth, alpha=0.7)
if s0 > thresh:
ax.plot(x0,
y0,
'.',
color=colors[l],
markersize=markersize,
alpha=0.7)
if s1 > thresh:
ax.plot(x1,
y1,
'.',
color=colors[l],
markersize=markersize,
alpha=0.7)
# for better visualization, add mid shoulder / mid hip
mid_shoulder = (
pnt[coco_keypoint_names[human_id].index('right_shoulder'), :2] +
pnt[coco_keypoint_names[human_id].index('left_shoulder'), :2]) / 2
mid_shoulder_sc = np.minimum(
pnt_sc[coco_keypoint_names[human_id].index('right_shoulder')],
pnt_sc[coco_keypoint_names[human_id].index('left_shoulder')])
mid_hip = (
pnt[coco_keypoint_names[human_id].index('right_hip'), :2] +
pnt[coco_keypoint_names[human_id].index('left_hip'), :2]) / 2
mid_hip_sc = np.minimum(
pnt_sc[coco_keypoint_names[human_id].index('right_hip')],
pnt_sc[coco_keypoint_names[human_id].index('left_hip')])
if (mid_shoulder_sc > thresh and
pnt_sc[coco_keypoint_names[human_id].index('nose')] > thresh):
y = [
mid_shoulder[0],
pnt[coco_keypoint_names[human_id].index('nose'), 0]
]
x = [
mid_shoulder[1],
pnt[coco_keypoint_names[human_id].index('nose'), 1]
]
line = ax.plot(x, y)
plt.setp(line,
color=colors[len(coco_point_skeleton)],
linewidth=linewidth,
alpha=0.7)
if (mid_shoulder_sc > thresh and mid_hip_sc > thresh):
y = [mid_shoulder[0], mid_hip[0]]
x = [mid_shoulder[1], mid_hip[1]]
line = ax.plot(x, y)
plt.setp(line,
color=colors[len(coco_point_skeleton) + 1],
linewidth=linewidth,
alpha=0.7)
return ax
def vis_bbox(img, bbox, label=None, score=None, label_names=None, ax=None):
"""Visualize bounding boxes inside image.
Example:
>>> from chainercv.datasets import VOCDetectionDataset
>>> from chainercv.datasets import voc_bbox_label_names
>>> from chainercv.visualizations import vis_bbox
>>> import matplotlib.pyplot as plot
>>> dataset = VOCDetectionDataset()
>>> img, bbox, label = dataset[60]
>>> vis_bbox(img, bbox, label,
... label_names=voc_bbox_label_names)
>>> plot.show()
Args:
img (~numpy.ndarray): An array of shape :math:`(3, height, width)`.
This is in RGB format and the range of its value is
:math:`[0, 255]`.
bbox (~numpy.ndarray): An array of shape :math:`(R, 4)`, where
:math:`R` is the number of bounding boxes in the image.
Each element is organized
by :math:`(y_{min}, x_{min}, y_{max}, x_{max})` in the second axis.
label (~numpy.ndarray): An integer array of shape :math:`(R,)`.
The values correspond to id for label names stored in
:obj:`label_names`. This is optional.
score (~numpy.ndarray): A float array of shape :math:`(R,)`.
Each value indicates how confident the prediction is.
This is optional.
label_names (iterable of strings): Name of labels ordered according
to label ids. If this is :obj:`None`, labels will be skipped.
ax (matplotlib.axes.Axis): The visualization is displayed on this
axis. If this is :obj:`None` (default), a new axis is created.
Returns:
~matploblib.axes.Axes:
Returns the Axes object with the plot for further tweaking.
"""
from matplotlib import pyplot as plot
if label is not None and not len(bbox) == len(label):
raise ValueError('The length of label must be same as that of bbox')
if score is not None and not len(bbox) == len(score):
raise ValueError('The length of score must be same as that of bbox')
# Returns newly instantiated matplotlib.axes.Axes object if ax is None
ax = vis_image(img, ax=ax)
# If there is no bounding box to display, visualize the image and exit.
if len(bbox) == 0:
return ax
for i, bb in enumerate(bbox):
xy = (bb[1], bb[0])
height = bb[2] - bb[0]
width = bb[3] - bb[1]
ax.add_patch(plot.Rectangle(
xy, width, height, fill=False, edgecolor='red', linewidth=3))
caption = []
if label is not None and label_names is not None:
lb = label[i]
if not (0 <= lb < len(label_names)):
raise ValueError('No corresponding name is given')
caption.append(label_names[lb])
if score is not None:
sc = score[i]
caption.append('{:.2f}'.format(sc))
if len(caption) > 0:
ax.text(bb[1], bb[0],
': '.join(caption),
style='italic',
bbox={'facecolor': 'white', 'alpha': 0.7, 'pad': 10})
return ax
def vis_bbox(img,
bbox,
label=None,
score=None,
mask=None,
label_names=None,
ax=None,
contour=False,
labeldisplay=True):
"""Visualize bounding boxes inside image.
Example:
>>> from chainercv.datasets import VOCDetectionDataset
>>> from chainercv.datasets import voc_bbox_label_names
>>> from chainercv.visualizations import vis_bbox
>>> import matplotlib.pyplot as plot
>>> dataset = VOCDetectionDataset()
>>> img, bbox, label = dataset[60]
>>> vis_bbox(img, bbox, label,
... label_names=voc_bbox_label_names)
>>> plot.show()
Args:
img (~numpy.ndarray): An array of shape :math:`(3, height, width)`.
This is in RGB format and the range of its value is
:math:`[0, 255]`.
bbox (~numpy.ndarray): An array of shape :math:`(R, 4)`, where
:math:`R` is the number of bounding boxes in the image.
Each element is organized
by :obj:`(y_min, x_min, y_max, x_max)` in the second axis.
label (~numpy.ndarray): An integer array of shape :math:`(R,)`.
The values correspond to id for label names stored in
:obj:`label_names`. This is optional.
score (~numpy.ndarray): A float array of shape :math:`(R,)`.
Each value indicates how confident the prediction is.
This is optional.
label_names (iterable of strings): Name of labels ordered according
to label ids. If this is :obj:`None`, labels will be skipped.
ax (matplotlib.axes.Axis): The visualization is displayed on this
axis. If this is :obj:`None` (default), a new axis is created.
Returns:
~matploblib.axes.Axes:
Returns the Axes object with the plot for further tweaking.
"""
from matplotlib import pyplot as plot
if label is not None and not len(bbox) == len(label):
raise ValueError('The length of label must be same as that of bbox')
if score is not None and not len(bbox) == len(score):
raise ValueError('The length of score must be same as that of bbox')
# alpha-blend the masks
COLOR = [(1, 1, 0), (1, 0, 1), (0, 1, 1), (0, 0, 1), (0, 1, 0), (1, 0, 0),
(0.1, 1, 0.2)]
dst = img.astype(float)
for i, m in enumerate(mask):
alpha = np.tile(np.round(m), (3, 1, 1)).astype(float) * 0.4
src1 = np.ones(dst.shape).astype(float)
for j, col in enumerate(COLOR[i % len(COLOR)]):
src1[j] *= col * 255
dst = cv2.multiply(src1, alpha) + cv2.multiply(dst, 1 - alpha)
# Returns newly instantiated matplotlib.axes.Axes object if ax is None
ax = vis_image(dst, ax=ax)
# If there is no bounding box to display, visualize the image and exit.
if len(bbox) == 0:
return ax
# add boxes, contours and labels
for i, bb in enumerate(bbox):
# boxes
xy = (bb[1], bb[0])
height = int(bb[2]) - int(bb[0])
width = int(bb[3]) - int(bb[1])
ax.add_patch(
plot.Rectangle(xy,
width,
height,
fill=False,
edgecolor='red',
linewidth=1))
# contours
if contour:
Mcontours = find_contours(mask[i].T, 0.5)
for verts in Mcontours:
p = Polygon(verts, facecolor="none", edgecolor=[0.5, 0.5, 0.5])
ax.add_patch(p)
#labels
caption = list()
if label is not None and label_names is not None:
lb = label[i]
print(lb)
if not (0 <= lb < len(label_names)):
raise ValueError('No corresponding name is given')
caption.append(label_names[lb])
if score is not None:
sc = score[i]
caption.append('{:.2f}'.format(sc))
if len(caption) > 0 and labeldisplay:
ax.text(bb[1],
bb[0],
': '.join(caption),
style='italic',
fontsize=8,
color='white'
) #'facecolor': 'white', 'alpha': 0.7, 'pad': 10})
return ax