Datasets:
Tasks:
Image-to-3D
Modalities:
Geospatial
Languages:
English
Tags:
gaussian-splatting
novel-view-synthesis
3d-reconstruction
semantic-segmentation
remote-sensing
drone-imagery
License:
File size: 4,663 Bytes
8d5dfff | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 | """
Creates a submission.zip with valid, correctly-sized and correctly-formatted
but randomly-generated content, so you can test your submission pipeline
(zip layout, image sizes, point cloud properties) before you have a trained
model. See this folder's README.md for the submission format this mirrors.
Only reads test/sparse/0/cameras.txt and images.txt from the dataset, so it
works for every scene, including final-testing scenes where the reference
photos and 3D ground truth are withheld.
Usage:
python make_dummy_submission.py --output submission.zip
"""
import argparse
import io
import zipfile
from pathlib import Path
import numpy as np
from PIL import Image
from plyfile import PlyData, PlyElement
DATASET_ROOT = Path(__file__).resolve().parent.parent # Twinworld_Datasets/
DATASETS = {
"tum": "Data_TUM",
"gold_coast": "Data_Goldcoast",
}
NUM_POINTS = 2000
SEED = 0
def read_cameras_txt(path):
"""Return {camera_id: (width, height)}."""
cameras = {}
with open(path) as f:
for line in f:
line = line.strip()
if not line or line.startswith('#'):
continue
parts = line.split()
camera_id = int(parts[0])
cameras[camera_id] = (int(parts[2]), int(parts[3]))
return cameras
def read_images_txt(path):
"""Return [(name, camera_id), ...]. Each image occupies two lines; the
second (POINTS2D) line is not needed here and is skipped."""
with open(path) as f:
lines = [line for line in f if not line.startswith('#')]
entries = []
for i in range(0, len(lines), 2):
parts = lines[i].split()
if not parts:
continue
entries.append((parts[9], int(parts[8])))
return entries
def scene_frames(scene_dir):
"""Return [(frame_stem, width, height), ...] for a scene's test poses."""
sparse_dir = scene_dir / "test" / "sparse" / "0"
cameras = read_cameras_txt(sparse_dir / "cameras.txt")
images = read_images_txt(sparse_dir / "images.txt")
return [(Path(name).stem, *cameras[camera_id]) for name, camera_id in images]
def random_png_bytes(width, height, rng):
pixels = rng.integers(0, 256, size=(height, width, 3), dtype=np.uint8)
buf = io.BytesIO()
Image.fromarray(pixels, mode="RGB").save(buf, format="PNG")
return buf.getvalue()
def random_point_cloud_bytes(num_points, rng, with_classification):
fields = [("x", "f4"), ("y", "f4"), ("z", "f4")]
if with_classification:
fields.append(("classification", "u1"))
vertex = np.empty(num_points, dtype=fields)
xyz = rng.uniform(-50.0, 50.0, size=(num_points, 3)).astype(np.float32)
vertex["x"], vertex["y"], vertex["z"] = xyz[:, 0], xyz[:, 1], xyz[:, 2]
if with_classification:
vertex["classification"] = rng.choice(
[0, 1, 2, 3, 4, 255], size=num_points).astype(np.uint8)
buf = io.BytesIO()
PlyData([PlyElement.describe(vertex, "vertex")], text=False, byte_order="<").write(buf)
return buf.getvalue()
def main():
parser = argparse.ArgumentParser(description=__doc__)
parser.add_argument("--output", type=Path, default=Path("submission.zip"),
help="output zip path")
parser.add_argument("--dataset_root", type=Path, default=DATASET_ROOT,
help="path to Twinworld_Datasets (containing Data_TUM/Data_Goldcoast)")
parser.add_argument("--num_points", type=int, default=NUM_POINTS,
help="random points per scene's point cloud")
parser.add_argument("--seed", type=int, default=SEED)
args = parser.parse_args()
rng = np.random.default_rng(args.seed)
n_scenes, n_frames = 0, 0
with zipfile.ZipFile(args.output, "w", zipfile.ZIP_DEFLATED) as zf:
for dataset_name, dir_name in DATASETS.items():
dataset_dir = args.dataset_root / dir_name
for scene_dir in sorted(dataset_dir.glob("scene_*")):
with_classification = dataset_name == "gold_coast"
for stem, width, height in scene_frames(scene_dir):
zf.writestr(f"{dataset_name}/{scene_dir.name}/rgb/{stem}.png",
random_png_bytes(width, height, rng))
n_frames += 1
ply_bytes = random_point_cloud_bytes(args.num_points, rng, with_classification)
zf.writestr(f"{dataset_name}/{scene_dir.name}/3D_point_cloud/point_cloud.ply",
ply_bytes)
n_scenes += 1
print(f"Wrote {args.output} ({n_scenes} scenes, {n_frames} rgb frames)")
if __name__ == "__main__":
main()
|