scene_camera = inout.load_scene_camera(scene_camera_path)
# Create folders for the output masks (if they do not exist yet).
mask_dir_path = os.path.dirname(dp_split['mask_tpath'].format(
scene_id=scene_id, im_id=0, gt_id=0))
misc.ensure_dir(mask_dir_path)
mask_visib_dir_path = os.path.dirname(dp_split['mask_visib_tpath'].format(
scene_id=scene_id, im_id=0, gt_id=0))
misc.ensure_dir(mask_visib_dir_path)
# Initialize a renderer.
misc.log('Initializing renderer...')
width, height = dp_split['im_size']
ren = renderer.create_renderer(width,
height,
renderer_type=p['renderer_type'],
mode='depth')
# Add object models.
for obj_id in dp_model['obj_ids']:
ren.add_object(obj_id, dp_model['model_tpath'].format(obj_id=obj_id))
im_ids = sorted(scene_gt.keys())
for im_id in im_ids:
if im_id % 100 == 0:
misc.log(
'Calculating masks - dataset: {} ({}, {}), scene: {}, im: {}'.
format(p['dataset'], p['dataset_split'],
p['dataset_split_type'], scene_id, im_id))
if p['scene_ids']:
scene_ids_curr = set(scene_ids_curr).intersection(p['scene_ids'])
# Rendering mode.
renderer_modalities = []
if p['vis_rgb']:
renderer_modalities.append('rgb')
if p['vis_depth_diff'] or (p['vis_rgb'] and p['vis_rgb_resolve_visib']):
renderer_modalities.append('depth')
renderer_mode = '+'.join(renderer_modalities)
# Create a renderer.
width, height = dp_split['im_size']
ren = renderer.create_renderer(width,
height,
p['renderer_type'],
mode=renderer_mode,
shading='flat')
# Load object models.
models = {}
for obj_id in dp_model['obj_ids']:
misc.log('Loading 3D model of object {}...'.format(obj_id))
model_path = dp_model['model_tpath'].format(obj_id=obj_id)
model_color = None
if not p['vis_orig_color']:
model_color = tuple(colors[(obj_id - 1) % len(colors)])
ren.add_object(obj_id, model_path, surf_color=model_color)
for scene_id in scene_ids_curr:
im_size_rgb = [int(round(x * float(ssaa_fact))) for x in dp_camera['im_size']]
K_rgb = dp_camera['K'] * ssaa_fact
# Intrinsic parameters for RGB rendering.
fx_rgb, fy_rgb, cx_rgb, cy_rgb =\
K_rgb[0, 0], K_rgb[1, 1], K_rgb[0, 2], K_rgb[1, 2]
# Intrinsic parameters for depth rendering.
K = dp_camera['K']
fx_d, fy_d, cx_d, cy_d = K[0, 0], K[1, 1], K[0, 2], K[1, 2]
# Create the RGB renderer.
width_rgb, height_rgb = im_size_rgb[0], im_size_rgb[1]
ren_rgb = renderer.create_renderer(width_rgb,
height_rgb,
renderer_type,
mode='rgb',
shading=shading)
ren_rgb.set_light_ambient_weight(ambient_weight)
# Add object models to the RGB renderer.
for obj_id in obj_ids:
ren_rgb.add_object(obj_id, dp_model['model_tpath'].format(obj_id=obj_id))
# Create the depth renderer.
width_depth, height_depth, = dp_camera['im_size'][0], dp_camera['im_size'][1]
ren_depth = renderer.create_renderer(width_depth,
height_depth,
renderer_type,
mode='depth')
model_type = 'eval'
dp_model = dataset_params.get_model_params(
p['datasets_path'], dataset, model_type)
# Rendering mode.
renderer_modalities = []
if p['vis_rgb']:
renderer_modalities.append('rgb')
if p['vis_depth_diff'] or (p['vis_rgb'] and p['vis_rgb_resolve_visib']):
renderer_modalities.append('depth')
renderer_mode = '+'.join(renderer_modalities)
# Create a renderer.
width, height = dp_split['im_size']
ren = renderer.create_renderer(
width, height, p['renderer_type'], mode=renderer_mode)
# Load object models.
models = {}
for obj_id in dp_model['obj_ids']:
misc.log('Loading 3D model of object {}...'.format(obj_id))
model_path = dp_model['model_tpath'].format(obj_id=obj_id)
model_color = None
if not p['vis_orig_color']:
model_color = tuple(colors[(obj_id - 1) % len(colors)])
ren.add_object(obj_id, model_path, surf_color=model_color)
# Load pose estimates.
misc.log('Loading pose estimates...')
ests = inout.load_bop_results(
os.path.join(config.results_path, result_fname))
model_type = None
if p['dataset'] == 'tless':
model_type = 'cad'
dp_model = dataset_params.get_model_params(p['datasets_path'], p['dataset'],
model_type)
# Initialize a renderer.
misc.log('Initializing renderer...')
# The renderer has a larger canvas for generation of masks of truncated objects.
im_width, im_height = dp_split['im_size']
ren_width, ren_height = 3 * im_width, 3 * im_height
ren_cx_offset, ren_cy_offset = im_width, im_height
ren = renderer.create_renderer(ren_width,
ren_height,
p['renderer_type'],
mode='depth')
for obj_id in dp_model['obj_ids']:
model_fpath = dp_model['model_tpath'].format(obj_id=obj_id)
ren.add_object(obj_id, model_fpath)
scene_ids = dataset_params.get_present_scene_ids(dp_split)
for scene_id in scene_ids:
# Load scene info and ground-truth poses.
scene_camera = inout.load_scene_camera(
dp_split['scene_camera_tpath'].format(scene_id=scene_id))
scene_gt = inout.load_scene_gt(
dp_split['scene_gt_tpath'].format(scene_id=scene_id))