chore: import upstream snapshot with attribution
This commit is contained in:
wehub-resource-sync
@@ -0,0 +1,98 @@
|
||||
<!--[metadata]
|
||||
title = "Open photogrammetry format"
|
||||
tags = ["2D", "3D", "Camera", "Photogrammetry"]
|
||||
thumbnail = "https://static.rerun.io/open-photogrammetry-format/c9bec43a3a3abd725a55ee8eb527a4c0cb01979b/480w.png"
|
||||
thumbnail_dimensions = [480, 480]
|
||||
channel = "release"
|
||||
include_in_manifest = true
|
||||
build_args = ["--jpeg-quality=50"]
|
||||
-->
|
||||
|
||||
Uses [`pyopf`](https://github.com/Pix4D/pyopf) to load and display a photogrammetrically reconstructed 3D point cloud in the [Open Photogrammetry Format (OPF)](https://www.pix4d.com/open-photogrammetry-format/).
|
||||
|
||||
<picture data-inline-viewer="examples/open_photogrammetry_format">
|
||||
<source media="(max-width: 480px)" srcset="https://static.rerun.io/open_photogrammetry_format/603d5605f9670889bc8bce3365f16b831fce1eb1/480w.png">
|
||||
<source media="(max-width: 768px)" srcset="https://static.rerun.io/open_photogrammetry_format/603d5605f9670889bc8bce3365f16b831fce1eb1/768w.png">
|
||||
<source media="(max-width: 1024px)" srcset="https://static.rerun.io/open_photogrammetry_format/603d5605f9670889bc8bce3365f16b831fce1eb1/1024w.png">
|
||||
<source media="(max-width: 1200px)" srcset="https://static.rerun.io/open_photogrammetry_format/603d5605f9670889bc8bce3365f16b831fce1eb1/1200w.png">
|
||||
<img src="https://static.rerun.io/open_photogrammetry_format/603d5605f9670889bc8bce3365f16b831fce1eb1/full.png" alt="">
|
||||
</picture>
|
||||
|
||||
## Used Rerun types
|
||||
[`Image`](https://www.rerun.io/docs/reference/types/archetypes/image), [`Points3D`](https://www.rerun.io/docs/reference/types/archetypes/points3d), [`Transform3D`](https://www.rerun.io/docs/reference/types/archetypes/transform3d), [`Pinhole`](https://www.rerun.io/docs/reference/types/archetypes/pinhole)
|
||||
|
||||
## Background
|
||||
|
||||
This example loads an Open Photogrammetry Format (OPF) project and displays the cameras and point cloud data.
|
||||
OPF, which stands for 'open photogrammetry format,' is a file format used for photogrammetry data.
|
||||
It contains all the necessary information related to a reconstructed 3D model made with photogrammetry, including calibration, point clouds and dense reconstruction.
|
||||
|
||||
## Logging and visualizing with Rerun
|
||||
|
||||
The visualizations in this example were created with the following Rerun code:
|
||||
|
||||
### Timelines
|
||||
|
||||
For each processed frame, all data sent to Rerun is associated with specific time using [`timelines`](https://www.rerun.io/docs/concepts/logging-and-ingestion/timelines).
|
||||
|
||||
```python
|
||||
rr.set_time("image", sequence=i)
|
||||
```
|
||||
|
||||
### Video
|
||||
|
||||
Pinhole camera is utilized for achieving a 3D view and camera perspective through the use of the [`Pinhole`](https://www.rerun.io/docs/reference/types/archetypes/pinhole) and [`Transform3D`](https://www.rerun.io/docs/reference/types/archetypes/transform3d) archetypes.
|
||||
|
||||
```python
|
||||
rr.log("world/cameras", rr.Transform3D(translation=calib_camera.position, mat3x3=rot))
|
||||
```
|
||||
|
||||
```python
|
||||
rr.log(
|
||||
"world/cameras/image",
|
||||
rr.Pinhole(
|
||||
resolution=sensor.image_size_px,
|
||||
focal_length=calib_sensor.internals.focal_length_px,
|
||||
principal_point=calib_sensor.internals.principal_point_px,
|
||||
camera_xyz=rr.ViewCoordinates.RUB,
|
||||
),
|
||||
)
|
||||
```
|
||||
The input video is logged as a sequence of [`Image`](https://www.rerun.io/docs/reference/types/archetypes/image) objects to the `world/cameras/image/rgb` entity.
|
||||
```python
|
||||
rr.log("world/cameras/image/rgb", rr.Image(np.array(img)).compress(jpeg_quality=jpeg_quality))
|
||||
```
|
||||
|
||||
### Point clouds
|
||||
|
||||
Point clouds from the project are logged as [`Points3D`](https://www.rerun.io/docs/reference/types/archetypes/points3d) archetype to the `world/points` entity.
|
||||
|
||||
```python
|
||||
rr.log("world/points", rr.Points3D(points.position, colors=points.color), static=True)
|
||||
```
|
||||
|
||||
|
||||
## Run the code
|
||||
|
||||
|
||||
> This example requires Python 3.10 or higher because of [`pyopf`](https://pypi.org/project/pyopf/).
|
||||
|
||||
To run this example, make sure you have the Rerun repository checked out and the latest SDK installed:
|
||||
```bash
|
||||
pip install --upgrade rerun-sdk # install the latest Rerun SDK
|
||||
git clone git@github.com:rerun-io/rerun.git # Clone the repository
|
||||
cd rerun
|
||||
git checkout latest # Check out the commit matching the latest SDK release
|
||||
```
|
||||
Install the necessary libraries specified in the requirements file:
|
||||
```bash
|
||||
pip install -e examples/python/open_photogrammetry_format
|
||||
```
|
||||
To experiment with the provided example, simply execute the main Python script:
|
||||
```bash
|
||||
python -m open_photogrammetry_format # run the example
|
||||
```
|
||||
If you wish to customize it or explore additional features, use the CLI with the `--help` option for guidance:
|
||||
```bash
|
||||
python -m open_photogrammetry_format --help
|
||||
```
|
||||
@@ -0,0 +1,239 @@
|
||||
#!/usr/bin/env python3
|
||||
"""
|
||||
Load an Open Photogrammetry Format (OPF) project and display the cameras and point cloud.
|
||||
|
||||
Requires Python 3.10 or higher because of [pyopf](https://pypi.org/project/pyopf/).
|
||||
"""
|
||||
|
||||
from __future__ import annotations
|
||||
|
||||
import argparse
|
||||
import logging
|
||||
import zipfile
|
||||
from dataclasses import dataclass
|
||||
from pathlib import Path
|
||||
from typing import Final
|
||||
|
||||
import numpy as np
|
||||
import requests
|
||||
import tqdm
|
||||
from PIL import Image
|
||||
from pyopf.io import load
|
||||
from pyopf.resolve import resolve
|
||||
|
||||
import rerun as rr
|
||||
|
||||
DESCRIPTION = """
|
||||
# Open Photogrammetry Format
|
||||
|
||||
Visualizes an Open Photogrammetry Format (OPF) project, displaying the cameras and point cloud.
|
||||
|
||||
The full source code for this example is available
|
||||
[on GitHub](https://github.com/rerun-io/rerun/blob/latest/examples/python/open_photogrammetry_format).
|
||||
|
||||
### Links
|
||||
* [OPF specification](https://pix4d.github.io/opf-spec/index.html)
|
||||
* [Dataset source](https://support.pix4d.com/hc/en-us/articles/360000235126#OPF)
|
||||
* [pyopf](https://github.com/Pix4D/pyopf)
|
||||
"""
|
||||
|
||||
|
||||
@dataclass
|
||||
class DatasetSpec:
|
||||
dir_name: str
|
||||
url: str
|
||||
|
||||
|
||||
DATASETS = {
|
||||
"olympic": DatasetSpec("olympic_flame", "https://data.pix4d.com/misc/example_datasets/olympic_flame.zip"),
|
||||
"rainwater": DatasetSpec(
|
||||
"catch_rainwater_demo",
|
||||
"https://data.pix4d.com/misc/example_datasets/catch_rainwater_demo.zip",
|
||||
),
|
||||
"rivaz": DatasetSpec("rivaz_demo", "https://data.pix4d.com/misc/example_datasets/rivaz_demo.zip"),
|
||||
}
|
||||
DATASET_DIR: Final = Path(__file__).parent / "dataset"
|
||||
|
||||
|
||||
def download_file(url: str, path: Path) -> None:
|
||||
path.parent.mkdir(parents=True, exist_ok=True)
|
||||
logging.info("Downloading %s to %s", url, path)
|
||||
response = requests.get(url, stream=True)
|
||||
with tqdm.tqdm.wrapattr(
|
||||
open(path, "wb"),
|
||||
"write",
|
||||
miniters=1,
|
||||
total=int(response.headers.get("content-length", 0)),
|
||||
desc=f"Downloading {path.name}",
|
||||
) as f:
|
||||
for chunk in response.iter_content(chunk_size=4096):
|
||||
f.write(chunk)
|
||||
|
||||
|
||||
def unzip_dir(archive: Path, destination: Path) -> None:
|
||||
"""Unzip the archive to the destination, using tqdm to display progress."""
|
||||
logging.info("Extracting %s to %s", archive, destination)
|
||||
with zipfile.ZipFile(archive, "r") as zip_ref:
|
||||
zip_ref.extractall(destination)
|
||||
|
||||
|
||||
class OPFProject:
|
||||
def __init__(self, path: Path, log_as_frames: bool = True) -> None:
|
||||
"""
|
||||
Create a new OPFProject from the given path.
|
||||
|
||||
Parameters
|
||||
----------
|
||||
path : Path
|
||||
Path to the project file.
|
||||
log_as_frames : bool, optional
|
||||
Whether to log the cameras as individual frames, by default True
|
||||
|
||||
"""
|
||||
self.path = path
|
||||
self.project = resolve(load(str(self.path)))
|
||||
self.log_as_frames = log_as_frames
|
||||
|
||||
@classmethod
|
||||
def from_dataset(cls, dataset: str, log_as_frames: bool = True) -> OPFProject:
|
||||
"""
|
||||
Download the dataset if necessary and return the project file.
|
||||
|
||||
Parameters
|
||||
----------
|
||||
dataset : str
|
||||
Name of the dataset to download.
|
||||
log_as_frames : bool, optional
|
||||
Whether to log the cameras as individual frames, by default True
|
||||
|
||||
"""
|
||||
spec = DATASETS[dataset]
|
||||
if not (DATASET_DIR / spec.dir_name).exists():
|
||||
zip_file = DATASET_DIR / f"{dataset}.zip"
|
||||
if not zip_file.exists():
|
||||
download_file(DATASETS[dataset].url, zip_file)
|
||||
unzip_dir(DATASET_DIR / f"{dataset}.zip", DATASET_DIR)
|
||||
|
||||
return cls(DATASET_DIR / spec.dir_name / "project.opf", log_as_frames=log_as_frames)
|
||||
|
||||
def log_point_cloud(self) -> None:
|
||||
"""Log the project's point cloud."""
|
||||
points = self.project.point_cloud_objs[0].nodes[0]
|
||||
rr.log("world/points", rr.Points3D(points.position, colors=points.color), static=True)
|
||||
|
||||
def log_calibrated_cameras(self, jpeg_quality: int | None) -> None:
|
||||
"""
|
||||
Log the project's calibrated cameras as individual frames.
|
||||
|
||||
Logging all cameras in a single frame is also possible, but clutter the default view with too many image views.
|
||||
"""
|
||||
sensor_map = {sensor.id: sensor for sensor in self.project.input_cameras.sensors}
|
||||
calib_sensor_map = {sensor.id: sensor for sensor in self.project.calibration.calibrated_cameras.sensors}
|
||||
|
||||
for i, (camera, calib_camera) in enumerate(
|
||||
zip(
|
||||
self.project.camera_list.cameras,
|
||||
self.project.calibration.calibrated_cameras.cameras,
|
||||
strict=False,
|
||||
),
|
||||
):
|
||||
if not str(camera.uri).endswith(".jpg"):
|
||||
continue
|
||||
|
||||
if self.log_as_frames:
|
||||
rr.set_time("image", sequence=i)
|
||||
entity = "world/cameras"
|
||||
else:
|
||||
entity = f"world/cameras/{i}"
|
||||
|
||||
sensor = sensor_map[calib_camera.sensor_id]
|
||||
calib_sensor = calib_sensor_map[calib_camera.sensor_id]
|
||||
|
||||
# Specification for the omega, phi, kappa angles:
|
||||
# https://pix4d.github.io/opf-spec/specification/calibrated_cameras.html#calibrated-camera
|
||||
omega, phi, kappa = tuple(np.deg2rad(a) for a in calib_camera.orientation_deg)
|
||||
rot = (
|
||||
np.array([
|
||||
[1, 0, 0],
|
||||
[0, np.cos(omega), -np.sin(omega)],
|
||||
[0, np.sin(omega), np.cos(omega)],
|
||||
])
|
||||
@ np.array([
|
||||
[np.cos(phi), 0, np.sin(phi)],
|
||||
[0, 1, 0],
|
||||
[-np.sin(phi), 0, np.cos(phi)],
|
||||
])
|
||||
@ np.array([
|
||||
[np.cos(kappa), -np.sin(kappa), 0],
|
||||
[np.sin(kappa), np.cos(kappa), 0],
|
||||
[0, 0, 1],
|
||||
])
|
||||
)
|
||||
|
||||
rr.log(entity, rr.Transform3D(translation=calib_camera.position, mat3x3=rot))
|
||||
|
||||
assert calib_sensor.internals.type == "perspective"
|
||||
|
||||
# RUB coordinate system specified in https://pix4d.github.io/opf-spec/specification/projected_input_cameras.html#coordinate-system-specification
|
||||
rr.log(
|
||||
entity + "/image",
|
||||
rr.Pinhole(
|
||||
resolution=sensor.image_size_px,
|
||||
focal_length=calib_sensor.internals.focal_length_px,
|
||||
principal_point=calib_sensor.internals.principal_point_px,
|
||||
camera_xyz=rr.ViewCoordinates.RUB,
|
||||
),
|
||||
)
|
||||
|
||||
if jpeg_quality is not None:
|
||||
with Image.open(self.path.parent / camera.uri) as img:
|
||||
rr.log(entity + "/image/rgb", rr.Image(img).compress(jpeg_quality=jpeg_quality))
|
||||
else:
|
||||
rr.log(entity + "/image/rgb", rr.EncodedImage(path=self.path.parent / camera.uri))
|
||||
|
||||
|
||||
def main() -> None:
|
||||
logging.getLogger().addHandler(rr.LoggingHandler())
|
||||
logging.getLogger().setLevel("INFO")
|
||||
|
||||
parser = argparse.ArgumentParser(
|
||||
description="Load an Open Photogrammetry Format (OPF) project and display the cameras and point cloud.",
|
||||
)
|
||||
parser.add_argument(
|
||||
"--dataset",
|
||||
choices=DATASETS.keys(),
|
||||
default="olympic",
|
||||
help="Run on a demo image automatically downloaded",
|
||||
)
|
||||
parser.add_argument(
|
||||
"--no-frames",
|
||||
action="store_true",
|
||||
help="Log all cameras globally instead of as individual frames in the timeline.",
|
||||
)
|
||||
parser.add_argument(
|
||||
"--jpeg-quality",
|
||||
type=int,
|
||||
default=None,
|
||||
help="If specified, compress the camera images with the given JPEG quality.",
|
||||
)
|
||||
|
||||
rr.script_add_args(parser)
|
||||
|
||||
args, unknown = parser.parse_known_args()
|
||||
for arg in unknown:
|
||||
logging.warning(f"unknown arg: {arg}")
|
||||
|
||||
# load the data set
|
||||
project = OPFProject.from_dataset(args.dataset, log_as_frames=not args.no_frames)
|
||||
|
||||
# display everything in Rerun
|
||||
rr.script_setup(args, "rerun_example_open_photogrammetry_format")
|
||||
rr.log("description", rr.TextDocument(DESCRIPTION, media_type=rr.MediaType.MARKDOWN), static=True)
|
||||
rr.log("world", rr.ViewCoordinates.RIGHT_HAND_Z_UP, static=True)
|
||||
project.log_point_cloud()
|
||||
project.log_calibrated_cameras(jpeg_quality=args.jpeg_quality)
|
||||
rr.script_teardown(args)
|
||||
|
||||
|
||||
if __name__ == "__main__":
|
||||
main()
|
||||
@@ -0,0 +1,13 @@
|
||||
[project]
|
||||
name = "open_photogrammetry_format"
|
||||
version = "0.1.0"
|
||||
requires-python = ">=3.10" # pyopf requirement
|
||||
readme = "README.md"
|
||||
dependencies = ["numpy", "pillow", "pyopf", "requests", "rerun-sdk", "tqdm"]
|
||||
|
||||
[project.scripts]
|
||||
open_photogrammetry_format = "open_photogrammetry_format:main"
|
||||
|
||||
[build-system]
|
||||
requires = ["hatchling"]
|
||||
build-backend = "hatchling.build"
|
||||
Reference in New Issue
Block a user