very very first commit, WIP git remote add origin [email protected]:cquest/sgblur.git

Author: cquest

ALGORITHMS.md

@@ -0,0 +1,60 @@
+# Blur algorithms
+
+GeoVisio blurring module offers various blur strategies. It can be selected using `STRATEGY` environment variable, which accepts `FAST`, `COMPROMISE`, `QUALITATIVE` and `LEGACY` values.
+
+If you use __Docker__, you may want to map container folder `/opt/blur/models` to an host folder where there is enough space to store blurring models.
+
+## Options
+
+Several specific options are available, and can be defined using environment variables.
+
+- `STRATEGY` (optional) : set algorithm to use for blurring, it depends on the level of speed and precision you require. Values are:
+  - `DISABLE` completely disables blurring
+  - `FAST` is the fastest algorithm but should be considered unreliable on pictures with lots of persons or large pictures like 360° images with persons in the background
+  - `COMPROMISE` should be reliable enough for all kinds of images but the blur may be jagged
+  - `QUALITATIVE` takes a lot more time to complete but will accomplish good detouring
+  - `LEGACY` theorically has the best results but is *way* slower than every other method (this is the blur algorithm used in GeoVisio versions <= 1.2.0)
+- `WEBP_METHOD` : quality/speed trade-off for WebP encoding of pictures derivates (0=fast, 6=slower-better, 6 by default).
+
+## Development notes
+
+This part presents more information about blur strategies implementation, this is a recommended-but-not-mandatory read 😉
+
+### General blur pipeline
+
+The blurring pipeline is as follow:
+
+1. Get an input image
+2. Split it into multiple quads, which sizes correspond to the object detection AI model (inferer)'s input size
+3. Give the quads to the inferer
+4. Get boxes that *may* contain cars, persons...
+5. Place them on a new single image, which size is as close as possible to the sementic segmentation AI model (segmenter)'s input size
+6. Give that image to the segmenter
+7. Get a blur mask back (containing a blur mask for each box)
+8. Arrange the blur masks back on the original image, given the inferer's boxes
+9. Apply the blur mask to the original image
+
+The inferer is [YOLOv6](https://github.com/meituan/YOLOv6) and the segmenter depends on the segmentation strategy.
+
+When using the `FAST` strategy, steps 2-5 and 8 are skipped and inferer is not used.
+
+On step 6 (no matter if steps 2-5 took place), if the image is larger than the segmenter's input size, it is reduced to fit in width *or* height and fed to the segmenter in one or more batches.
+
+### Logs
+
+There are several log messages comming from Tensorflow, Pytorch and YOLO that cannot be easily suppressed :\
+`INFO: Created TensorFlow Lite XNNPACK delegate for CPU.`, see [a relating issue](https://github.com/google/mediapipe/issues/2354). To fix, a Tensorflow build from source is required.\
+`...UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument...`, this is a YOLOv6 generated warning, see [a fix](https://github.com/pytorch/pytorch/issues/50276). To fix, add `, indexing='ij'` in the calls to `torch.meshgrid()` in files `effidehead.py` and `loss.py` of YOLOv6's source.
+
+### GPU/TPU usage
+
+To run the segmenter on the GPU, a CUDA capable GPU is required, run `pip uninstall tensorflow && pip install tensorflow-gpu`. For more information see [the Tensorflow GPU guide](https://www.tensorflow.org/guide/gpu).
+
+To run the inferer on the GPU, refer to the [Pytorch download page](https://pytorch.org/) to find the right Pytorch version with CUDA capabilities.
+Both the inferer and segmenter should automatically switch to GPU, this __hasn't been tested__ in a production environment. If you see any issues, please [let us know](https://gitlab.com/PanierAvide/geovisio/-/issues).
+
+For TPU usage, refer to the [Tensorflow tpu guide](https://www.tensorflow.org/guide/tpu), Pytorch seems to be compatible with TPUs but it remains to be verified whether YOLO's architectures also is.
+
+### On YOLOv6 updates
+
+Current GeoVisio code and documentation fixes YOLOv6 to its 0.2.0 release. This has been done because trained models are not compatible through versions, and should be explicitly changed in some parts of the code. When updating, make sure to update models download links and YOLOv6 release tag everywhere necessary in GeoVisio code.

CODE_OF_CONDUCT.md

@@ -0,0 +1,134 @@
+
+# Code of Conduct
+
+## Our Pledge
+
+We as members, contributors, and leaders pledge to make participation in our
+community a harassment-free experience for everyone, regardless of age, body
+size, visible or invisible disability, ethnicity, sex characteristics, gender
+identity and expression, level of experience, education, socio-economic status,
+nationality, personal appearance, race, caste, color, religion, or sexual
+identity and orientation.
+
+We pledge to act and interact in ways that contribute to an open, welcoming,
+diverse, inclusive, and healthy community.
+
+## Our Standards
+
+Examples of behavior that contributes to a positive environment for our
+community include:
+
+* Demonstrating empathy and kindness toward other people
+* Being respectful of differing opinions, viewpoints, and experiences
+* Giving and gracefully accepting constructive feedback
+* Accepting responsibility and apologizing to those affected by our mistakes,
+  and learning from the experience
+* Focusing on what is best not just for us as individuals, but for the overall
+  community
+
+Examples of unacceptable behavior include:
+
+* The use of sexualized language or imagery, and sexual attention or advances of
+  any kind
+* Trolling, insulting or derogatory comments, and personal or political attacks
+* Public or private harassment
+* Publishing others' private information, such as a physical or email address,
+  without their explicit permission
+* Other conduct which could reasonably be considered inappropriate in a
+  professional setting
+
+## Enforcement Responsibilities
+
+Community leaders are responsible for clarifying and enforcing our standards of
+acceptable behavior and will take appropriate and fair corrective action in
+response to any behavior that they deem inappropriate, threatening, offensive,
+or harmful.
+
+Community leaders have the right and responsibility to remove, edit, or reject
+comments, commits, code, wiki edits, issues, and other contributions that are
+not aligned to this Code of Conduct, and will communicate reasons for moderation
+decisions when appropriate.
+
+## Scope
+
+This Code of Conduct applies within all community spaces, and also applies when
+an individual is officially representing the community in public spaces.
+Examples of representing our community include using an official e-mail address,
+posting via an official social media account, or acting as an appointed
+representative at an online or offline event.
+
+## Enforcement
+
+Instances of abusive, harassing, or otherwise unacceptable behavior may be
+reported to the community leaders responsible for enforcement at [email protected]
+
+All complaints will be reviewed and investigated promptly and fairly.
+
+All community leaders are obligated to respect the privacy and security of the
+reporter of any incident.
+
+## Enforcement Guidelines
+
+Community leaders will follow these Community Impact Guidelines in determining
+the consequences for any action they deem in violation of this Code of Conduct:
+
+### 1. Correction
+
+**Community Impact**: Use of inappropriate language or other behavior deemed
+unprofessional or unwelcome in the community.
+
+**Consequence**: A private, written warning from community leaders, providing
+clarity around the nature of the violation and an explanation of why the
+behavior was inappropriate. A public apology may be requested.
+
+### 2. Warning
+
+**Community Impact**: A violation through a single incident or series of
+actions.
+
+**Consequence**: A warning with consequences for continued behavior. No
+interaction with the people involved, including unsolicited interaction with
+those enforcing the Code of Conduct, for a specified period of time. This
+includes avoiding interactions in community spaces as well as external channels
+like social media. Violating these terms may lead to a temporary or permanent
+ban.
+
+### 3. Temporary Ban
+
+**Community Impact**: A serious violation of community standards, including
+sustained inappropriate behavior.
+
+**Consequence**: A temporary ban from any sort of interaction or public
+communication with the community for a specified period of time. No public or
+private interaction with the people involved, including unsolicited interaction
+with those enforcing the Code of Conduct, is allowed during this period.
+Violating these terms may lead to a permanent ban.
+
+### 4. Permanent Ban
+
+**Community Impact**: Demonstrating a pattern of violation of community
+standards, including sustained inappropriate behavior, harassment of an
+individual, or aggression toward or disparagement of classes of individuals.
+
+**Consequence**: A permanent ban from any sort of public interaction within the
+community.
+
+## Attribution
+
+This Code of Conduct is adapted from the [Contributor Covenant][homepage],
+version 2.1, available at
+[https://www.contributor-covenant.org/version/2/1/code_of_conduct.html][v2.1].
+
+Community Impact Guidelines were inspired by
+[Mozilla's code of conduct enforcement ladder][Mozilla CoC].
+
+For answers to common questions about this code of conduct, see the FAQ at
+[https://www.contributor-covenant.org/faq][FAQ]. Translations are available at
+[https://www.contributor-covenant.org/translations][translations].
+
+[homepage]: https://www.contributor-covenant.org
+[v2.1]: https://www.contributor-covenant.org/version/2/1/code_of_conduct.html
+[Mozilla CoC]: https://github.com/mozilla/diversity
+[FAQ]: https://www.contributor-covenant.org/faq
+[translations]: https://www.contributor-covenant.org/translations
+

LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2022 Adrien Pavie
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

README.md

@@ -0,0 +1,150 @@
+# GeoVisio Blurring Algorithms
+
+![GeoVisio logo](https://gitlab.com/geovisio/api/-/blob/52df05fd9c95f1a929b32bee9735505eeeddc7e8/images/logo_full.png)
+
+[GeoVisio](https://gitlab.com/geovisio) is a complete solution for storing and __serving your own geolocated pictures__ (like [StreetView](https://www.google.com/streetview/) / [Mapillary](https://mapillary.com/)).
+
+This repository only contains __the blurring algorithms and API__. This repository can be used completely independently from others. [All other components are listed here](https://gitlab.com/geovisio).
+
+
+## Install
+
+### System dependencies
+
+Some algorithms (compromise and qualitative) will need system dependencies :
+
+- ffmpeg
+- libsm6
+- libxext6
+
+You can install them through your package manager, for example in Ubuntu:
+
+```bash
+sudo apt install ffmpeg libsm6 libxext6
+```
+
+### Retrieve code
+
+You can download code from this repository with git clone:
+
+```bash
+git clone https://gitlab.com/geovisio/blurring.git
+cd blurring/
+```
+
+### Other dependencies
+
+We use [Git Submodules](https://git-scm.com/book/en/v2/Git-Tools-Submodules) to manage some of our dependencies. Run the following command to get these dependencies:
+
+```bash
+git submodule update --init
+```
+
+We also use Pip to handle Python dependencies. You can create a virtual environment first:
+
+```bash
+python -m venv env
+source ./env/bin/activate
+```
+
+And depending on if you want to use API, command-line scripts or both, run these commands:
+
+```bash
+pip install -r requirements-bin.txt  # For CLI
+pip install -r requirements-api.txt  # For API
+```
+
+If at some point you're lost or need help, you can contact us through [issues](https://gitlab.com/geovisio/blurring/-/issues) or by [email](mailto:[email protected]).
+
+
+## Usage
+
+### Command-line interface
+
+All details of available commands are listed in [USAGE.md](./USAGE.md) documentation, or by calling this command:
+
+```bash
+python src/main.py --help
+```
+
+A single picture can be blurred using the following command:
+
+```bash
+python src/main.py <path the the picture> <path to the output picture>
+```
+
+You can also launch the CLI through Docker:
+
+```bash
+docker run \
+	geovisio/blurring \
+	cli
+```
+
+### Web API
+
+The Web API can be launched with the following command:
+
+```bash
+uvicorn src.api:app --reload
+```
+
+It is then accessible on [localhost:8000](http://127.0.0.1:8000).
+
+You can also launch the API through Docker:
+
+```bash
+docker run \
+	-p 8000:80 \
+	--name geovisio_blurring \
+	geovisio/blurring \
+	api
+```
+
+API documentation is available under `/docs` route, so [localhost:8000/docs](http://127.0.0.1:8000/docs) if you use local instance.
+
+A single picture can be blurred using the following HTTP call (here made using _curl_):
+
+```bash
+# Considering your picture is called my_picture.jpg
+curl -X 'POST' \
+  'http://127.0.0.1:8000/blur/' \
+  -H 'accept: image/webp' \
+  -H 'Content-Type: multipart/form-data' \
+  -F 'picture=@my_picture.jpg;type=image/jpeg' \
+  --output blurred.webp
+```
+
+Note that various settings can be changed to control API behaviour. You can edit them [using one of the described method in FastAPI documentation](https://fastapi.tiangolo.com/advanced/settings/). Available settings are:
+
+- `STRATEGY`: blur algorithm to use (FAST, LEGACY, COMPROMISE, QUALITATIVE)
+- `WEBP_METHOD`: quality/speed trade-off for WebP encoding of pictures derivates (0=fast, 6=slower-better, 6 by default)
+
+
+## Contributing
+
+Pull requests are welcome. For major changes, please open an issue first to discuss what you would like to change.
+
+You might want to read more [about available blur algorithms](./ALGORITHMS.md).
+
+### Testing
+
+Tests are handled with Pytest. You can run them using:
+
+```bash
+pip install -r requirements-dev.txt
+pytest
+```
+
+### Documentation
+
+High-level documentation for command-line script is handled by [Typer](https://typer.tiangolo.com/). You can update the generated `USAGE.md` file using this command:
+
+```bash
+make docs
+```
+
+
+## License
+
+Copyright (c) GeoVisio team 2022-2023, [released under MIT license](./LICENSE).

USAGE.md

@@ -0,0 +1,22 @@
+# `python src/main.py`
+
+GeoVisio blurring scripts
+
+**Usage**:
+
+```console
+$ python src/main.py [OPTIONS] INPUT OUTPUT COMMAND [ARGS]...
+```
+
+**Arguments**:
+
+* `INPUT`: Picture to blur  [required]
+* `OUTPUT`: Output file path  [required]
+
+**Options**:
+
+* `--strategy [fast|legacy|compromise|qualitative]`: Blur algorithm to use  [default: Strategy.fast]
+* `--mask / --picture`: Get a blur mask instead of blurred picture  [default: picture]
+* `--install-completion`: Install completion for the current shell.
+* `--show-completion`: Show completion for the current shell, to copy it or customize the installation.
+* `--help`: Show this message and exit.

models/yolov8s_panoramax.pt

@@ -0,0 +1,3 @@
+ultralytics==8.0.58
+pyturbojpeg==1.7.0
+Pillow-simd