|
from multiprocessing import Pool |
|
from tqdm import tqdm |
|
from pathlib import Path |
|
import numpy as np |
|
from collections import deque |
|
import argparse |
|
import cv2 |
|
|
|
def get_raycast_building_mask(building_grid): |
|
laser_range = 200 |
|
num_laser = 100 |
|
robot_pos = (building_grid.shape[0] // 2-1, building_grid.shape[1] // 2 - 1) |
|
unoccupied_pos = np.stack(np.where(building_grid != 1), axis=1) |
|
|
|
if len(unoccupied_pos) == 0: |
|
return None |
|
|
|
l2_dist = unoccupied_pos - [robot_pos[0], robot_pos[1]] |
|
closest = ((l2_dist ** 2).sum(1)**0.5).argmin() |
|
|
|
robot_pos = (unoccupied_pos[closest][0], unoccupied_pos[closest][1]) |
|
|
|
free_points, hit_points, actual_hit_points = get_free_points_in_front(building_grid, robot_pos, laser_range=laser_range, num_laser=num_laser) |
|
free_points[:, 0][free_points[:, 0] >= building_grid.shape[0]] = building_grid.shape[0] - 1 |
|
free_points[:, 1][free_points[:, 1] >= building_grid.shape[1]] = building_grid.shape[1] - 1 |
|
free_points[:, 0][free_points[:, 0] < 0] = 0 |
|
free_points[:, 1][free_points[:, 1] < 0] = 0 |
|
|
|
hit_points[:, 0][hit_points[:, 0] >= building_grid.shape[0]] = building_grid.shape[0] - 1 |
|
hit_points[:, 1][hit_points[:, 1] >= building_grid.shape[1]] = building_grid.shape[1] - 1 |
|
hit_points[:, 0][hit_points[:, 0] < 0] = 0 |
|
hit_points[:, 1][hit_points[:, 1] < 0] = 0 |
|
|
|
if len(free_points) > 0: |
|
|
|
|
|
inited_flood_grid = init_flood_fill(robot_pos, hit_points, building_grid.shape) |
|
inited_flood_grid = (inited_flood_grid * 255).astype(np.uint8).copy() |
|
|
|
|
|
np.random.shuffle(free_points) |
|
|
|
for i in range(len(free_points)): |
|
seed_point = free_points[i] |
|
if inited_flood_grid[seed_point[0], seed_point[1]] != 0: |
|
break |
|
else: |
|
print('Unable to find a valid seed point') |
|
return None |
|
|
|
num_filled, flooded_image, mask, bounding_box = cv2.floodFill(inited_flood_grid.copy(), None, seedPoint=(seed_point[1], seed_point[0]), newVal=0) |
|
|
|
return flooded_image |
|
else: |
|
print("No free points") |
|
return None |
|
|
|
def flood_fill_simple(center_point, occupancy_map): |
|
""" |
|
center_point: starting point (x,y) of fill |
|
occupancy_map: occupancy map generated from Bresenham ray-tracing |
|
""" |
|
|
|
occupancy_map = np.copy(occupancy_map) |
|
sx, sy = occupancy_map.shape |
|
fringe = deque() |
|
fringe.appendleft(center_point) |
|
while fringe: |
|
|
|
n = fringe.pop() |
|
nx, ny = n |
|
unknown_val = 0.5 |
|
|
|
if nx > 0: |
|
if occupancy_map[nx - 1, ny] == unknown_val: |
|
occupancy_map[nx - 1, ny] = 0 |
|
fringe.appendleft((nx - 1, ny)) |
|
|
|
if nx < sx - 1: |
|
if occupancy_map[nx + 1, ny] == unknown_val: |
|
occupancy_map[nx + 1, ny] = 0 |
|
fringe.appendleft((nx + 1, ny)) |
|
|
|
if ny > 0: |
|
if occupancy_map[nx, ny - 1] == unknown_val: |
|
occupancy_map[nx, ny - 1] = 0 |
|
fringe.appendleft((nx, ny - 1)) |
|
|
|
if ny < sy - 1: |
|
if occupancy_map[nx, ny + 1] == unknown_val: |
|
occupancy_map[nx, ny + 1] = 0 |
|
fringe.appendleft((nx, ny + 1)) |
|
return occupancy_map |
|
|
|
def init_flood_fill(robot_pos, obstacle_points, occ_grid_shape): |
|
""" |
|
center_point: center point |
|
obstacle_points: detected obstacles points (x,y) |
|
xy_points: (x,y) point pairs |
|
""" |
|
center_x, center_y = robot_pos |
|
prev_ix, prev_iy = center_x, center_y |
|
occupancy_map = (np.ones(occ_grid_shape)) * 0.5 |
|
|
|
obstacle_points = np.vstack((obstacle_points, obstacle_points[0])) |
|
for (x, y) in zip(obstacle_points[:,0], obstacle_points[:,1]): |
|
|
|
ix = int(x) |
|
|
|
iy = int(y) |
|
free_area = bresenham((prev_ix, prev_iy), (ix, iy)) |
|
for fa in free_area: |
|
occupancy_map[fa[0]][fa[1]] = 0 |
|
prev_ix = ix |
|
prev_iy = iy |
|
return occupancy_map |
|
|
|
show_animation = False |
|
|
|
def bresenham(start, end): |
|
""" |
|
Implementation of Bresenham's line drawing algorithm |
|
See en.wikipedia.org/wiki/Bresenham's_line_algorithm |
|
Bresenham's Line Algorithm |
|
Produces a np.array from start and end (original from roguebasin.com) |
|
>>> points1 = bresenham((4, 4), (6, 10)) |
|
>>> print(points1) |
|
np.array([[4,4], [4,5], [5,6], [5,7], [5,8], [6,9], [6,10]]) |
|
""" |
|
|
|
x1, y1 = start |
|
x2, y2 = end |
|
dx = x2 - x1 |
|
dy = y2 - y1 |
|
is_steep = abs(dy) > abs(dx) |
|
if is_steep: |
|
x1, y1 = y1, x1 |
|
x2, y2 = y2, x2 |
|
|
|
swapped = False |
|
if x1 > x2: |
|
x1, x2 = x2, x1 |
|
y1, y2 = y2, y1 |
|
swapped = True |
|
dx = x2 - x1 |
|
dy = y2 - y1 |
|
error = int(dx / 2.0) |
|
y_step = 1 if y1 < y2 else -1 |
|
|
|
y = y1 |
|
points = [] |
|
for x in range(x1, x2 + 1): |
|
coord = [y, x] if is_steep else (x, y) |
|
points.append(coord) |
|
error -= abs(dy) |
|
if error < 0: |
|
y += y_step |
|
error += dx |
|
if swapped: |
|
points.reverse() |
|
points = np.array(points) |
|
return points |
|
|
|
def get_free_points_in_front(occupancy_grid, robot_pos, laser_range=10, num_laser=100): |
|
""" |
|
Assumes circular lidar |
|
occupancy_grid: np.array (h x w) |
|
robot_pos: (x, y) |
|
|
|
Outputs: |
|
free_points: np.array of hit points (x, y) |
|
""" |
|
|
|
free_points = [] |
|
hit_points = [] |
|
actual_hit_points = [] |
|
for orientation in np.linspace(np.pi/2, 3*np.pi/2, num_laser): |
|
end_point = (round(robot_pos[0] + laser_range * np.cos(orientation)), round(robot_pos[1] + laser_range * np.sin(orientation))) |
|
|
|
|
|
bresenham_points = (bresenham(robot_pos, end_point)) |
|
|
|
|
|
|
|
for i in range(len(bresenham_points)): |
|
|
|
if bresenham_points[i,0] < 0 or bresenham_points[i,0] >= occupancy_grid.shape[0] or bresenham_points[i,1] < 0 or bresenham_points[i,1] >= occupancy_grid.shape[1]: |
|
if i != 0: |
|
hit_points.append(bresenham_points[i-1]) |
|
break |
|
|
|
if occupancy_grid[bresenham_points[i,0], bresenham_points[i,1]] == 1: |
|
|
|
for j in range(min(4, len(bresenham_points) - i - 1)): |
|
free_points.append(bresenham_points[i+j]) |
|
|
|
actual_hit_points.append(bresenham_points[i + j + 1]) |
|
hit_points.append(bresenham_points[i + j + 1]) |
|
|
|
break |
|
else: |
|
free_point = bresenham_points[i] |
|
free_points.append(free_point) |
|
|
|
if i == len(bresenham_points) - 1: |
|
hit_points.append(end_point) |
|
break |
|
|
|
|
|
|
|
free_points = np.array(free_points) |
|
hit_points = np.array(hit_points) |
|
actual_hit_points = np.array(actual_hit_points) |
|
return free_points, hit_points, actual_hit_points |
|
|
|
if __name__ == "__main__": |
|
|
|
parser = argparse.ArgumentParser() |
|
parser.add_argument("--dataset_folder", type=str, default="/path/to/raycast") |
|
parser.add_argument("--class_idx_building", type=int, default=4) |
|
parser.add_argument("--num_workers", type=int, default=60) |
|
parser.add_argument("--location", type=str, default="los_angeles") |
|
|
|
args = parser.parse_args() |
|
|
|
dataset_folder = Path(args.dataset_folder) |
|
bev_folder = dataset_folder / args.location / "semantic_masks" |
|
output_folder = dataset_folder / args.location / "flood_fill" |
|
|
|
output_folder.mkdir(exist_ok=True, parents=True) |
|
|
|
def generate_mask(filepath): |
|
mask = np.load(filepath) |
|
building_grid = mask[..., args.class_idx_building] |
|
try: |
|
flooded_image = get_raycast_building_mask(building_grid) |
|
except: |
|
raise Exception(f"Error in {filepath}") |
|
|
|
if flooded_image is not None: |
|
output_file = output_folder / filepath.name |
|
np.save(output_file, flooded_image) |
|
else: |
|
print("No flood fill generated") |
|
|
|
bev_files = list(bev_folder.iterdir()) |
|
|
|
with Pool(args.num_workers) as p: |
|
for _ in tqdm(p.imap_unordered(generate_mask, bev_files), total=len(bev_files)): |
|
pass |
|
|
|
|