def savefig():
visualize.display_instances(image, gt_bbox, gt_mask, gt_class_ids,
[categories.category2name(i) for i in range(categories.cate_cnt)],
savefilename=os.path.join(save_visual_path, '%05d_gt.jpg' % i))
visualize.display_instances(image, bbox, mask, class_ids,
[categories.category2name(i) for i in range(categories.cate_cnt)],
savefilename=os.path.join(save_visual_path, '%05d_pred.jpg' % i))
def load_s3d(self, subset):
assert subset in ['train', 'test', 'val']
file_dict = dataset_tool.load_fileinfo()
for i in range(1, categories.cate_cnt):
self.add_class("s3d", i, categories.category2name(i))
if subset == 'train':
limit = [0, 3000]
elif subset == 'val':
limit = [3000, 3250]
else:
limit = [3250, 3500]
count = 0
for scene in file_dict:
if not (scene >= limit[0] and scene < limit[1]):
continue
for room in file_dict[scene]:
for position in file_dict[scene][room]:
count += 1
full_path = os.path.join(dataset_path,
'scene_%05d' % scene,
'2D_rendering', str(room),
'perspective', 'full',
str(position))
self.add_image(source="s3d",
image_id=count,
path=self.full_path(scene, room, position),
width=IMAGE_WIDTH,
height=IMAGE_HEIGHT,
scene=scene,
room=room,
position=position)
