def visualize_predictions(self, datum, output, iteration):
# FIXME: Refactor visualization code.
coords = datum['coords'].numpy()
batch_size = coords[:, 0].max() + 1
output_path = pathlib.Path(self.config.visualize_path)
output_path.mkdir(exist_ok=True)
for i in range(batch_size):
# Visualize RGB input.
coords_mask = coords[:, 0] == i
coords_b = coords[coords_mask, 1:]
if datum['input'].shape[1] > 3:
rgb_b = ((datum['input'][:, :3] + 0.5) * 255).numpy()
else:
rgb_b = ((datum['input'].repeat(1, 3) - datum['input'].min())
/ (datum['input'].max() - datum['input'].min()) * 255).numpy()
ptc = np.hstack((coords_b, rgb_b))
bbox_scores = output['detection'][i][:, -1]
bbox_mask = bbox_scores > self.config.visualize_min_confidence
if self.dataset.IS_ROTATION_BBOX:
bboxes = pc_utils.visualize_bboxes(output['detection'][i][bbox_mask][:, :7],
output['detection'][i][bbox_mask][:, 7],
bbox_param='xyzxyzr')
else:
bboxes = pc_utils.visualize_bboxes(output['detection'][i][bbox_mask][:, :6],
output['detection'][i][bbox_mask][:, 6])
pc_utils.visualize_pcd(ptc, bboxes)
scene_f = SCANNET_ALIGNMENT_PATH % (filename, filename)
alignment_txt = [
l for l in read_txt(scene_f) if l.startswith('axisAlignment = ')
][0]
rot = np.array([float(x)
for x in alignment_txt[16:].split()]).reshape(4, 4)
mesh.transform(rot)
pred_ours = np.load(os.path.join(OURS_PRED_PATH, 'out_%03d.npy.npz' % i))
gt = pc_utils.visualize_bboxes(pred_ours['gt'][:, :6],
pred_ours['gt'][:, 6],
num_points=NUM_BBOX_POINTS)
pred_ours = pc_utils.visualize_bboxes(pred_ours['pred'][:, :6],
pred_ours['pred'][:, 6],
num_points=NUM_BBOX_POINTS)
params = pc_utils.visualize_pcd(gt,
mesh,
save_image=f'viz_{filename}_gt.png')
pc_utils.visualize_pcd(mesh,
camera_params=params,
save_image=f'viz_{filename}_input.png')
pc_utils.visualize_pcd(pred_ours,
mesh,
camera_params=params,
save_image=f'viz_{filename}_ours.png')
pred_votenet = np.load(os.path.join(VOTENET_PRED_PATH, f'{filename}.npy'),
allow_pickle=True)[0]
votenet_preds = []
for pred_cls, pred_bbox, pred_score in pred_votenet:
pred_bbox = pred_bbox[:, (0, 2, 1)]
pred_bbox[:, -1] *= -1
votenet_preds.append(
import os
import numpy as np
import lib.pc_utils as pc_utils
from lib.utils import read_txt
OURS_PRED_PATH = 'outputs/visualization/ours_stanford'
STANFORD_PATH = '/cvgl/group/Stanford3dDataset_v1.2/Area_5'
NUM_BBOX_POINTS = 1000
files = sorted(read_txt('/scr/jgwak/Datasets/stanford3d/test.txt'))
for i, fn in enumerate(files):
area_name = os.path.splitext(fn.split(os.sep)[-1])[0]
area_name = '_'.join(area_name.split('_')[1:])[:-2]
# area_name = 'office_34'
# i = [i for i, fn in enumerate(files) if area_name in fn][0]
if area_name.startswith('WC_') or area_name.startswith('hallway_'):
continue
ptc_fn = os.path.join(STANFORD_PATH, area_name, f'{area_name}.txt')
ptc = np.array([l.split() for l in read_txt(ptc_fn)]).astype(float)
pred_ours = np.load(os.path.join(OURS_PRED_PATH, 'out_%03d.npy.npz' % i))
gt = pc_utils.visualize_bboxes(pred_ours['gt'][:, :6], pred_ours['gt'][:, 6],
num_points=NUM_BBOX_POINTS)
pred_ours = pc_utils.visualize_bboxes(pred_ours['pred'][:, :6], pred_ours['pred'][:, 6],
num_points=NUM_BBOX_POINTS)
params = pc_utils.visualize_pcd(gt, ptc, save_image=f'viz_Area5_{area_name}_gt.png')
pc_utils.visualize_pcd(ptc, camera_params=params, save_image=f'viz_Area5_{area_name}_input.png')
pc_utils.visualize_pcd(pred_ours, ptc, camera_params=params,
save_image=f'viz_Area5_{area_name}_ours.png')
def main():
pcd = o3d.io.read_point_cloud(INPUT_PCD)
pcd_xyz, pcd_feats = np.asarray(pcd.points), np.asarray(pcd.colors)
print(f'Finished reading {INPUT_PCD}:')
print(f'# points: {pcd_xyz.shape[0]} points')
print(f'volume: {np.prod(pcd_xyz.max(0) - pcd_xyz.min(0))} m^3')
sparse_voxelizer = Voxelizer(voxel_size=0.05)
height = pcd_xyz[:, LOCFEAT_IDX].copy()
height -= np.percentile(height, 0.99)
pcd_feats = np.hstack((pcd_feats, height[:, None]))
preprocess = []
for i in range(7):
start = time.time()
coords, feats, labels, transformation = sparse_voxelizer.voxelize(
pcd_xyz, pcd_feats, None)
preprocess.append(time.time() - start)
print('Voxelization time average: ', np.mean(preprocess[2:]))
coords = ME.utils.batched_coordinates([torch.from_numpy(coords).int()])
feats = torch.from_numpy(feats.astype(np.float32)).to('cuda')
config = get_config()
DatasetClass = load_dataset(config.dataset)
dataloader = initialize_data_loader(DatasetClass,
config,
threads=config.threads,
phase=config.test_phase,
augment_data=False,
shuffle=False,
repeat=False,
batch_size=config.test_batch_size,
limit_numpoints=False)
pipeline_model = load_pipeline(config, dataloader.dataset)
if config.weights.lower() != 'none':
state = torch.load(config.weights)
pipeline_model.load_state_dict(
state['state_dict'], strict=(not config.lenient_weight_loading))
pipeline_model.eval()
evaltime = []
for i in range(7):
start = time.time()
sinput = ME.SparseTensor(feats, coords).to('cuda')
datum = {'sinput': sinput, 'anchor_match_coords': None}
outputs = pipeline_model(datum, False)
evaltime.append(time.time() - start)
print('Network runtime average: ', np.mean(evaltime[2:]))
pred = outputs['detection'][0]
pred_mask = pred[:, -1] > MIN_CONF
pred = pred[pred_mask]
print(f'Detected {pred.shape[0]} instances')
bbox_xyz = pred[:, :6]
bbox_xyz += 0.5
bbox_xyz[:, :3] += 0.5
bbox_xyz[:, 3:] -= 0.5
bbox_xyz[:, 3:] = np.maximum(bbox_xyz[:, 3:], bbox_xyz[:, :3] + 0.1)
bbox_xyz = bbox_xyz.reshape(-1, 3)
bbox_xyz1 = np.hstack((bbox_xyz, np.ones((bbox_xyz.shape[0], 1))))
bbox_xyz = np.linalg.solve(transformation.reshape(4, 4),
bbox_xyz1.T).T[:, :3].reshape(-1, 6)
pred = np.hstack((bbox_xyz, pred[:, 6:]))
pred_pcd = pc_utils.visualize_bboxes(pred[:, :6],
pred[:, 6],
num_points=1000)
mesh = o3d.io.read_triangle_mesh(INPUT_MESH)
mesh.compute_vertex_normals()
pc_utils.visualize_pcd(mesh, pred_pcd)