fr_cam_rec = nusc.get('sample_data',
sample_rec['data']['CAM_FRONT_RIGHT'])
fl_cam_rec = nusc.get('sample_data',
sample_rec['data']['CAM_FRONT_LEFT'])
br_cam_rec = nusc.get('sample_data',
sample_rec['data']['CAM_BACK_RIGHT'])
bl_cam_rec = nusc.get('sample_data',
sample_rec['data']['CAM_BACK_LEFT'])
# ann_recs = [nusc.get('sample_annotation', token) for token in sample_rec['anns']]
has_more_frames = True
while has_more_frames:
## FRONT
impath, boxes, camera_intrinsic = nusc.get_sample_data(
f_cam_rec['token'])
points_f, coloring_f, im1 = nusc.explorer.map_pointcloud_to_image(
f_rad_rec['token'], f_cam_rec['token'])
points_fr, coloring_fr, im2 = nusc.explorer.map_pointcloud_to_image(
fr_rad_rec['token'], f_cam_rec['token'])
points_fl, coloring_fl, im3 = nusc.explorer.map_pointcloud_to_image(
fl_rad_rec['token'], f_cam_rec['token'])
print(points_fr)
print(points_fl)
ax1 = fig.add_subplot(1, 2, 1)
ax2 = fig.add_subplot(1, 2, 2)
# ax1.imshow(img1)
ax1.imshow(im1)
# nusc.render_sample_data(f_cam_rec['token'], True, ax=ax1)
ax1.scatter(points_f[0, :], points_f[1, :], c=coloring_f, s=5)
scene_rec = nusc.get('scene', scene['token'])
sample_rec = nusc.get('sample', scene_rec['first_sample_token'])
f_rad_rec = nusc.get('sample_data', sample_rec['data']['RADAR_FRONT'])
f_cam_rec = nusc.get('sample_data', sample_rec['data']['CAM_FRONT'])
has_more_frames = True
while has_more_frames:
bboxes = []
ax1 = fig.add_subplot(1,1,1)
# ax2 = fig.add_subplot(1,2,2)
camera_token = f_cam_rec['token']
radar_token = f_rad_rec['token']
## FRONT CAM + RADAR
impath, boxes, camera_intrinsic = nusc.get_sample_data(camera_token)
points, coloring, image = nusc.explorer.map_pointcloud_to_image(radar_token,
camera_token)
points[2, :] = coloring
# Plot the 3D boxes and create 2D COCO bboxes
# for box in boxes:
# coco_cat, cat_id, coco_supercat = nuscene_cat_to_coco(box.name)
# if coco_cat is None:
# continue
# c = np.array(nusc.explorer.get_color(box.name)) / 255.0
# box.render(ax1, view=camera_intrinsic, normalize=True, colors=[c, c, c])
# bboxes.append(box.to_coco_bbox(camera_intrinsic, image.size))
# # Plot the 3D boxes
# nusc.render_sample_data(camera_token, ax=ax1)
output_path = 'tmp/'
# iterate over all 3977 samples
for idx, sample in enumerate(nusc.sample):
sample_token = sample['token']
sample_record = nusc.get('sample', sample_token)
selected_data = {}
for channel, sample_data_token in sample_record['data'].items():
sample_data_record = nusc.get('sample_data', sample_data_token)
sensor_modality = sample_data_record['sensor_modality']
if sensor_modality in ['lidar']:
# create output directory
os.makedirs(output_path + channel, exist_ok=True)
selected_data[channel] = sample_data_token
# boxes returned are transformed from global to camera/LIDAR space within get_sample_data
data_path, boxes, camera_intrinsic = nusc.get_sample_data(
sample_data_token, box_vis_level=BoxVisibility.IN_FRONT)
with open(output_path + channel + '/' + str(idx).zfill(6) + '.txt',
'w') as file_writer:
for box in boxes:
category = box.name
truncated = 0
occluded = 0
alpha = 0
width = box.wlh[0]
length = box.wlh[1]
height = box.wlh[2]
posx = box.center[0]
posy = box.center[1]
posz = box.center[2]
rotation_y = box.orientation.angle
xmin = 0