本文整理汇总了Python中pycocotools.coco.COCO.decodeMask方法的典型用法代码示例。如果您正苦于以下问题：Python COCO.decodeMask方法的具体用法？Python COCO.decodeMask怎么用？Python COCO.decodeMask使用的例子？那么恭喜您, 这里精选的方法代码示例或许可以为您提供帮助。您也可以进一步了解该方法所在类pycocotools.coco.COCO的用法示例。
在下文中一共展示了COCO.decodeMask方法的1个代码示例，这些例子默认根据受欢迎程度排序。您可以为喜欢或者感觉有用的代码点赞，您的评价将有助于系统推荐出更棒的Python代码示例。
示例1: __init__
# 需要导入模块: from pycocotools.coco import COCO [as 别名]
# 或者: from pycocotools.coco.COCO import decodeMask [as 别名]
#.........这里部分代码省略.........
self.assign_value(factors)
print('salient value computed.!')
def factored_size(self,ann):
im_sz = {}
for item in ann:
sz = round(item['area']/self.size_norm,2)
im_sz[item['id']] = sz
return im_sz
def factored_loc(self,ann):
im_loc = {}
for item in ann:
loc = item['bbox']
loc = round(loc[0]+loc[2]/2)+640*(round(loc[1]+loc[3]/2)-1)
loc = loc/self.size_norm
im_loc[item['id']] = loc
return im_loc
def factored_den(self,ann):
im_den = {}
distmat = []
for item in ann:
coord = item['bbox']
c_coord = [coord[0]+coord[2]/2,coord[1]+coord[3]/2]
distmat.append(c_coord)
for item1,item2 in zip(distmat,ann):
xa = [item1]
den = sp.distance.cdist(xa,distmat)
den = den.mean()
im_den[item2['id']] = den
return im_den
def factored_dtc(self,ann):
im_dtc = {}
c = [320,240]
for item in ann:
coord = item['bbox']
c_coord = [coord[0]+coord[2]/2,coord[1]+coord[3]/2]
d = sp.distance.pdist([c,c_coord])
d = d[0]
d = d*2/self.loc_norm
im_dtc[item['id']] = d
return im_dtc
def save_saldict_tomatfile(self,sal_dict_to_save,name):
datalist = []
for item in sal_dict_to_save.keys():
im_sal = sal_dict_to_save[item]
saveitem = []
for im_item in im_sal.keys():
saveitem.append([im_item,im_sal[im_item]])
datalist.append(saveitem)
sio.savemat(name,{'sal_data':datalist})
def plot_saliencymap(self,saliency_dict,image_id):
im_id_indataset = self.SALICON['SALICON_id'][image_id]
ann_IDlist = self.Ins_coco.getAnnIds(im_id_indataset)
ann_list = self.Ins_coco.loadAnns(ann_IDlist)
sal_dict = saliency_dict[image_id]
maxv = max(sal_dict.values())
blankim = np.zeros((480,640,3),np.uint8)
plt.imshow(blankim)
ax = plt.gca()
polygons = []
color = []
for item in ann_list:
c =sal_dict[item['id']]/(maxv)
c = [c,c,c]
if type(item['segmentation']) == list:
# polygon
for seg in item['segmentation']:
poly = np.array(seg).reshape((len(seg)/2, 2))
polygons.append(Polygon(poly, True,alpha=0.4))
color.append(c)
else:
# mask
mask = self.Ins_coco.decodeMask(item['segmentation'])
img = np.ones( (mask.shape[0], mask.shape[1], 3) )
color_mask = c
# if ann['iscrowd'] == 1:
# color_mask = np.array([2.0,166.0,101.0])/255
# if ann['iscrowd'] == 0:
for i in range(3):
img[:,:,i] = color_mask[i]
ax.imshow(np.dstack( (img, mask*0.5) ))
p = PatchCollection(polygons, facecolors=color, edgecolors=(0,0,0,1), linewidths=0.5, alpha=0.9)
ax.add_collection(p)