def run(config): print("refine rough registration of fragments.") o3d.utility.set_verbosity_level(o3d.utility.VerbosityLevel.Debug) ply_file_names = get_file_list( join(config["path_dataset"], config["folder_fragment"]), ".ply") make_posegraph_for_refined_scene(ply_file_names, config) optimize_posegraph_for_refined_scene(config["path_dataset"], config) path_dataset = config['path_dataset'] n_fragments = len(ply_file_names) # Save to trajectory poses = [] pose_graph_fragment = o3d.io.read_pose_graph( join(path_dataset, config["template_refined_posegraph_optimized"])) for fragment_id in range(len(pose_graph_fragment.nodes)): pose_graph_rgbd = o3d.io.read_pose_graph( join(path_dataset, config["template_fragment_posegraph_optimized"] % fragment_id)) for frame_id in range(len(pose_graph_rgbd.nodes)): frame_id_abs = fragment_id * \ config['n_frames_per_fragment'] + frame_id pose = np.dot(pose_graph_fragment.nodes[fragment_id].pose, pose_graph_rgbd.nodes[frame_id].pose) poses.append(pose) traj_name = join(path_dataset, config["template_global_traj"]) write_poses_to_log(traj_name, poses)
def run(config): print("register fragments.") o3d.utility.set_verbosity_level(o3d.utility.VerbosityLevel.Debug) ply_file_names = get_file_list( join(config["path_dataset"], config["folder_fragment"]), ".ply") make_clean_folder(join(config["path_dataset"], config["folder_scene"])) make_posegraph_for_scene(ply_file_names, config) optimize_posegraph_for_scene(config["path_dataset"], config)
def make_pointcloud_for_fragment(path_dataset, color_files, depth_files, fragment_id, n_fragments, intrinsic, config): mesh = integrate_rgb_frames_for_fragment( color_files, depth_files, fragment_id, n_fragments, join(path_dataset, config["template_fragment_posegraph_optimized"] % fragment_id), intrinsic, config) pcd = o3d.geometry.PointCloud() pcd.points = mesh.vertices pcd.colors = mesh.vertex_colors pcd_name = join(path_dataset, config["template_fragment_pointcloud"] % fragment_id) o3d.io.write_point_cloud(pcd_name, pcd, False, True)
def make_posegraph_for_refined_scene(ply_file_names, config): pose_graph = o3d.io.read_pose_graph( join(config["path_dataset"], config["template_global_posegraph_optimized"])) n_files = len(ply_file_names) matching_results = {} for edge in pose_graph.edges: s = edge.source_node_id t = edge.target_node_id matching_results[s * n_files + t] = \ matching_result(s, t, edge.transformation) if config["python_multi_threading"] == True: from joblib import Parallel, delayed import multiprocessing import subprocess MAX_THREAD = min(multiprocessing.cpu_count(), max(len(pose_graph.edges), 1)) results = Parallel(n_jobs=MAX_THREAD)( delayed(register_point_cloud_pair)( ply_file_names, matching_results[r].s, matching_results[r].t, matching_results[r].transformation, config) for r in matching_results) for i, r in enumerate(matching_results): matching_results[r].transformation = results[i][0] matching_results[r].information = results[i][1] else: for r in matching_results: (matching_results[r].transformation, matching_results[r].information) = \ register_point_cloud_pair(ply_file_names, matching_results[r].s, matching_results[r].t, matching_results[r].transformation, config) pose_graph_new = o3d.pipelines.registration.PoseGraph() odometry = np.identity(4) pose_graph_new.nodes.append( o3d.pipelines.registration.PoseGraphNode(odometry)) for r in matching_results: (odometry, pose_graph_new) = update_posegraph_for_scene( matching_results[r].s, matching_results[r].t, matching_results[r].transformation, matching_results[r].information, odometry, pose_graph_new) print(pose_graph_new) o3d.io.write_pose_graph( join(config["path_dataset"], config["template_refined_posegraph"]), pose_graph_new)
def compute_initial_registration(s, t, source_down, target_down, source_fpfh, target_fpfh, path_dataset, config): if t == s + 1: # odometry case print("Using RGBD odometry") pose_graph_frag = o3d.io.read_pose_graph( join(path_dataset, config["template_fragment_posegraph_optimized"] % s)) n_nodes = len(pose_graph_frag.nodes) transformation_init = np.linalg.inv(pose_graph_frag.nodes[n_nodes - 1].pose) (transformation, information) = \ multiscale_icp(source_down, target_down, [config["voxel_size"]], [50], config, transformation_init) else: # loop closure case (success, transformation, information) = register_point_cloud_fpfh(source_down, target_down, source_fpfh, target_fpfh, config) if not success: print("No reasonable solution. Skip this pair") return (False, np.identity(4), np.zeros((6, 6))) print(transformation) if config["debug_mode"]: draw_registration_result(source_down, target_down, transformation) return (True, transformation, information)
def make_posegraph_for_fragment(path_dataset, sid, eid, color_files, depth_files, fragment_id, n_fragments, intrinsic, with_opencv, config): o3d.utility.set_verbosity_level(o3d.utility.VerbosityLevel.Error) pose_graph = o3d.pipelines.registration.PoseGraph() trans_odometry = np.identity(4) pose_graph.nodes.append( o3d.pipelines.registration.PoseGraphNode(trans_odometry)) for s in range(sid, eid): for t in range(s + 1, eid): # odometry if t == s + 1: print( "Fragment %03d / %03d :: RGBD matching between frame : %d and %d" % (fragment_id, n_fragments - 1, s, t)) [success, trans, info] = register_one_rgbd_pair(s, t, color_files, depth_files, intrinsic, with_opencv, config) trans_odometry = np.dot(trans, trans_odometry) trans_odometry_inv = np.linalg.inv(trans_odometry) pose_graph.nodes.append( o3d.pipelines.registration.PoseGraphNode( trans_odometry_inv)) pose_graph.edges.append( o3d.pipelines.registration.PoseGraphEdge(s - sid, t - sid, trans, info, uncertain=False)) # keyframe loop closure if s % config['n_keyframes_per_n_frame'] == 0 \ and t % config['n_keyframes_per_n_frame'] == 0: print( "Fragment %03d / %03d :: RGBD matching between frame : %d and %d" % (fragment_id, n_fragments - 1, s, t)) [success, trans, info] = register_one_rgbd_pair(s, t, color_files, depth_files, intrinsic, with_opencv, config) if success: pose_graph.edges.append( o3d.pipelines.registration.PoseGraphEdge( s - sid, t - sid, trans, info, uncertain=True)) o3d.io.write_pose_graph( join(path_dataset, config["template_fragment_posegraph"] % fragment_id), pose_graph)
def make_posegraph_for_scene(ply_file_names, config): pose_graph = o3d.pipelines.registration.PoseGraph() odometry = np.identity(4) pose_graph.nodes.append(o3d.pipelines.registration.PoseGraphNode(odometry)) n_files = len(ply_file_names) matching_results = {} for s in range(n_files): for t in range(s + 1, n_files): matching_results[s * n_files + t] = matching_result(s, t) if config["python_multi_threading"] == True: from joblib import Parallel, delayed import multiprocessing import subprocess MAX_THREAD = min(multiprocessing.cpu_count(), max(len(matching_results), 1)) results = Parallel(n_jobs=MAX_THREAD)(delayed( register_point_cloud_pair)(ply_file_names, matching_results[r].s, matching_results[r].t, config) for r in matching_results) for i, r in enumerate(matching_results): matching_results[r].success = results[i][0] matching_results[r].transformation = results[i][1] matching_results[r].information = results[i][2] else: for r in matching_results: (matching_results[r].success, matching_results[r].transformation, matching_results[r].information) = \ register_point_cloud_pair(ply_file_names, matching_results[r].s, matching_results[r].t, config) for r in matching_results: if matching_results[r].success: (odometry, pose_graph) = update_posegraph_for_scene( matching_results[r].s, matching_results[r].t, matching_results[r].transformation, matching_results[r].information, odometry, pose_graph) o3d.io.write_pose_graph( join(config["path_dataset"], config["template_global_posegraph"]), pose_graph)
def run(config): print("making fragments from RGBD sequence.") make_clean_folder(join(config["path_dataset"], config["folder_fragment"])) [color_files, depth_files] = get_rgbd_file_lists(config["path_dataset"]) n_files = len(color_files) n_fragments = int( math.ceil(float(n_files) / config['n_frames_per_fragment'])) if config["python_multi_threading"] is True: from joblib import Parallel, delayed import multiprocessing import subprocess MAX_THREAD = min(multiprocessing.cpu_count(), n_fragments) Parallel(n_jobs=MAX_THREAD)(delayed(process_single_fragment) (fragment_id, color_files, depth_files, n_files, n_fragments, config) for fragment_id in range(n_fragments)) else: for fragment_id in range(n_fragments): process_single_fragment(fragment_id, color_files, depth_files, n_files, n_fragments, config)
def run(config): print("slac non-rigid optimization.") o3d.utility.set_verbosity_level(o3d.utility.VerbosityLevel.Debug) # dataset path and slac subfolder path # slac default subfolder for 0.050 voxel size: `dataset/slac/0.050/`. path_dataset = config["path_dataset"] slac_folder = join(path_dataset, config["subfolder_slac"]) # Read RGBD images. [color_files, depth_files] = get_rgbd_file_lists(config["path_dataset"]) if len(color_files) != len(depth_files): raise ValueError( "The number of color images {} must equal to the number of depth images {}." .format(len(color_files), len(depth_files))) # Read optimized pose graph. [Generated by `register` stage]. posegraph = o3d.io.read_pose_graph( join(slac_folder, config["template_optimized_posegraph_slac"])) # If camera intrinsic is not provided, # the default PrimeSense intrinsic is used. if config["path_intrinsic"]: intrinsic = o3d.io.read_pinhole_camera_intrinsic( config["path_intrinsic"]) else: intrinsic = o3d.camera.PinholeCameraIntrinsic( o3d.camera.PinholeCameraIntrinsicParameters.PrimeSenseDefault) focal_length = intrinsic.get_focal_length() principal_point = intrinsic.get_principal_point() intrinsic_t = o3d.core.Tensor([[focal_length[0], 0, principal_point[0]], [0, focal_length[1], principal_point[1]], [0, 0, 1]]) device = o3d.core.Device( 'CUDA:0' if o3d.core.cuda.is_available() else 'CPU:0') voxel_grid = o3d.t.geometry.VoxelBlockGrid( attr_names=('tsdf', 'weight', 'color'), attr_dtypes=(o3c.float32, o3c.float32, o3c.float32), attr_channels=((1), (1), (3)), voxel_size=config['tsdf_cubic_size'] / 512, block_resolution=16, block_count=config['block_count'], device=device) # Load control grid. ctr_grid_keys = o3d.core.Tensor.load(slac_folder + "ctr_grid_keys.npy") ctr_grid_values = o3d.core.Tensor.load(slac_folder + "ctr_grid_values.npy") ctr_grid = o3d.t.pipelines.slac.control_grid(3.0 / 8, ctr_grid_keys.to(device), ctr_grid_values.to(device), device) fragment_folder = join(path_dataset, config["folder_fragment"]) k = 0 depth_scale = float(config['depth_scale']) depth_max = float(config['max_depth']) for i in range(len(posegraph.nodes)): fragment_pose_graph = o3d.io.read_pose_graph( join(fragment_folder, "fragment_optimized_%03d.json" % i)) for node in fragment_pose_graph.nodes: pose_local = node.pose extrinsic_local_t = o3d.core.Tensor(np.linalg.inv(pose_local)) pose = np.dot(posegraph.nodes[i].pose, node.pose) extrinsic_t = o3d.core.Tensor(np.linalg.inv(pose)) depth = o3d.t.io.read_image(depth_files[k]).to(device) color = o3d.t.io.read_image(color_files[k]).to(device) rgbd = o3d.t.geometry.RGBDImage(color, depth) print('Deforming and integrating Frame {:3d}'.format(k)) rgbd_projected = ctr_grid.deform(rgbd, intrinsic_t, extrinsic_local_t, depth_scale, depth_max) frustum_block_coords = voxel_grid.compute_unique_block_coordinates( rgbd_projected.depth, intrinsic_t, extrinsic_t, depth_scale, depth_max) voxel_grid.integrate(frustum_block_coords, rgbd_projected.depth, rgbd_projected.color, intrinsic_t, extrinsic_t, depth_scale, depth_max) k = k + 1 if (config["save_output_as"] == "pointcloud"): pcd = voxel_grid.extract_point_cloud().to(o3d.core.Device("CPU:0")) save_pcd_path = join(slac_folder, "output_slac_pointcloud.ply") o3d.t.io.write_point_cloud(save_pcd_path, pcd) else: mesh = voxel_grid.extract_triangle_mesh().to(o3d.core.Device("CPU:0")) mesh_legacy = mesh.to_legacy() save_mesh_path = join(slac_folder, "output_slac_mesh.ply") o3d.io.write_triangle_mesh(save_mesh_path, mesh_legacy)
def run(config): print("slac non-rigid optimization.") o3d.utility.set_verbosity_level(o3d.utility.VerbosityLevel.Debug) path_dataset = config['path_dataset'] ply_file_names = get_file_list( join(config["path_dataset"], config["folder_fragment"]), ".ply") if (len(ply_file_names) == 0): raise RuntimeError( "No fragment found in {}, please make sure the reconstruction_system has finished running on the dataset." .format(join(config["path_dataset"], config["folder_fragment"]))) pose_graph_fragment = o3d.io.read_pose_graph( join(path_dataset, config["template_refined_posegraph_optimized"])) # SLAC optimizer parameters. slac_params = o3d.t.pipelines.slac.slac_optimizer_params( max_iterations=config["max_iterations"], voxel_size=config["voxel_size"], distance_threshold=config["distance_threshold"], fitness_threshold=config["fitness_threshold"], regularizer_weight=config["regularizer_weight"], device=o3d.core.Device(str(config["device"])), slac_folder=join(path_dataset, config["folder_slac"])) # SLAC debug option. debug_option = o3d.t.pipelines.slac.slac_debug_option(False, 0) # Run the system. pose_graph_updated = o3d.pipelines.registration.PoseGraph() # rigid optimization method. if (config["method"] == "rigid"): pose_graph_updated = o3d.t.pipelines.slac.run_rigid_optimizer_for_fragments( ply_file_names, pose_graph_fragment, slac_params, debug_option) elif (config["method"] == "slac"): pose_graph_updated, ctrl_grid = o3d.t.pipelines.slac.run_slac_optimizer_for_fragments( ply_file_names, pose_graph_fragment, slac_params, debug_option) hashmap = ctrl_grid.get_hashmap() active_buf_indices = hashmap.active_buf_indices().to( o3d.core.Dtype.Int64) key_tensor = hashmap.key_tensor()[active_buf_indices] key_tensor.save( join(slac_params.get_subfolder_name(), "ctr_grid_keys.npy")) value_tensor = hashmap.value_tensor()[active_buf_indices] value_tensor.save( join(slac_params.get_subfolder_name(), "ctr_grid_values.npy")) else: raise RuntimeError( "Requested optimization method {}, is not implemented. Implemented methods includes slac and rigid." .format(config["method"])) # Write updated pose graph. o3d.io.write_pose_graph( join(slac_params.get_subfolder_name(), config["template_optimized_posegraph_slac"]), pose_graph_updated) # Write trajectory for slac-integrate stage. fragment_folder = join(path_dataset, config["folder_fragment"]) params = [] for i in range(len(pose_graph_updated.nodes)): fragment_pose_graph = o3d.io.read_pose_graph( join(fragment_folder, "fragment_optimized_%03d.json" % i)) for node in fragment_pose_graph.nodes: pose = np.dot(pose_graph_updated.nodes[i].pose, node.pose) param = o3d.camera.PinholeCameraParameters() param.extrinsic = np.linalg.inv(pose) params.append(param) trajectory = o3d.camera.PinholeCameraTrajectory() trajectory.parameters = params o3d.io.write_pinhole_camera_trajectory( slac_params.get_subfolder_name() + "/optimized_trajectory_" + str(config["method"]) + ".log", trajectory)