diff --git a/README.md b/README.md index b8287e94..9678eb5f 100644 --- a/README.md +++ b/README.md @@ -8,7 +8,7 @@ [![Code style: black](https://img.shields.io/badge/code%20style-black-000000.svg)](https://github.com/psf/black) -Toolbox and trackers for object pose-estimation. Based on the work [CosyPose](https://github.com/Simple-Robotics/cosypose) and [MegaPose](https://github.com/megapose6d/megapose6d). This directory is currently under development. +Toolbox and trackers for object pose-estimation. Based on the work [CosyPose](https://github.com/Simple-Robotics/cosypose) and [MegaPose](https://github.com/megapose6d/megapose6d). This directory is currently under development. Please refer to the [documentation](https://agimus-project.github.io/happypose/) for more details. # Installation @@ -34,3 +34,44 @@ cd happypose/pose_estimators/megapose/deps/bop_toolkit_challenge/ # Remove all versions enforcing on requirements.txt pip install -r requirements.txt -e . ``` + + +# Create data directory + +``` +Create data dir /somewhere/convenient. The dataset to store are quite large. +export MEGAPOSE_DATA_DIR=/somewhere/convenient + +``` + +# Configuration for the evaluation + +If you plan on evaluating CosyPose and Megapose, you need to modify the following lines in `bop_toolkit_lib/config.py`, replace : + +``` +######## Basic ######## + + +# Folder with the BOP datasets. +if 'BOP_PATH' in os.environ: + datasets_path = os.environ['BOP_PATH'] +else: + datasets_path = r'/path/to/bop/datasets' + +# Folder with pose results to be evaluated. +results_path = r'/path/to/folder/with/results' + +# Folder for the calculated pose errors and performance scores. +eval_path = r'/path/to/eval/folder' +``` + +with + +``` +######## Basic ######## + +# Folder with the BOP datasets. +datasets_path = str(os.environ['BOP_DATASETS_PATH']) +results_path = str(os.environ['BOP_RESULTS_PATH']) +eval_path = str(os.environ['BOP_EVAL_PATH']) +``` \ No newline at end of file diff --git a/docs/book/SUMMARY.md b/docs/book/SUMMARY.md index a1d6a72b..d9daaf12 100644 --- a/docs/book/SUMMARY.md +++ b/docs/book/SUMMARY.md @@ -11,4 +11,6 @@ # MegaPose -- [TODO](./megapose/todo.md) +- [Testing your installation](./megapose/test-install.md) +- [Downloading and preparing the data](./megapose/download-data.md) +- [Evaluate MegaPose](./megapose/evaluate.md) diff --git a/docs/book/cosypose/download-data.md b/docs/book/cosypose/download-data.md index 047a4d74..09ad99ad 100644 --- a/docs/book/cosypose/download-data.md +++ b/docs/book/cosypose/download-data.md @@ -1,15 +1,7 @@ # Downloading and preparing the data -``` -Create data dir /somewhere/convenient. The dataset to store are quite large. -export MEGAPOSE_DATA_DIR=/somewhere/convenient -``` - -download [barbecue sauce](https://drive.google.com/drive/folders/10BIvhnrKGbNr8EKGB3KUtkSNcp460k9S) and put it in `$MEGAPOSE_DATA_DIR/examples/barbecue-sauce/` - - -All data used (datasets, models, results, ...) are stored in a directory `local_data` at the root of the repository. Create it with `mkdir local_data` or use a symlink if you want the data to be stored at a different place. We provide the utility `cosypose/scripts/download.py` for downloading required data and models. All of the files can also be [downloaded manually](https://drive.google.com/drive/folders/1JmOYbu1oqN81Dlj2lh6NCAMrC8pEdAtD?usp=sharing). +All data used (datasets, models, results, ...) are stored in a directory `$MEGAPOSE_DATA_DIR` that you created in the Readsme. We provide the utilities for downloading required data and models. All of the files can also be [downloaded manually](https://www.paris.inria.fr/archive_ylabbeprojectsdata/). ## BOP Datasets @@ -111,16 +103,4 @@ Notes: - The PoseCNN detections (and coarse pose estimates) on YCB-Video were extracted and converted from [these PoseCNN results](https://github.com/yuxng/YCB_Video_toolbox/blob/master/results_PoseCNN_RSS2018.zip). - The Pix2pose detections were extracted using [pix2pose's](https://github.com/kirumang/Pix2Pose) code. We used the detection model from their paper, see [here](https://github.com/kirumang/Pix2Pose#download-pre-trained-weights). For the ViVo detections, their code was slightly modified. The code used to extract detections can be found [here](https://github.com/ylabbe/pix2pose_cosypose). - - -
-Megapose - - ## 1. Download pre-trained pose estimation models -Download pose estimation models to $MEGAPOSE_DATA_DIR/megapose-models: - -``` -python -m happypose.toolbox.utils.download --megapose_models -``` -
- + \ No newline at end of file diff --git a/docs/book/cosypose/evaluate.md b/docs/book/cosypose/evaluate.md index 9e519724..ec6ea23a 100644 --- a/docs/book/cosypose/evaluate.md +++ b/docs/book/cosypose/evaluate.md @@ -1,84 +1,6 @@ # Evaluating CosyPose -## Install - -Two installation steps are needed : [bop_toolkit](https://github.com/thodan/bop_toolkit) and [bop_renderer](https://github.com/thodan/bop_renderer/). These repository are stored in `happypose/pose_estimators/megapose/deps/`. - -### 1. Bop_toolkit_challenge - -``` -cd /happypose/pose_estimators/megapose/deps/bop_toolkit_challenge -# You need to remove all the versions from the requirements.txt file, then : -pip install -r requirements.txt -e . -``` - -Then, you need to modify the following lines in `bop_toolkit_lib/config.py`, replace : - - -``` -######## Basic ######## - - -# Folder with the BOP datasets. -if 'BOP_PATH' in os.environ: - datasets_path = os.environ['BOP_PATH'] -else: - datasets_path = r'/path/to/bop/datasets' - -# Folder with pose results to be evaluated. -results_path = r'/path/to/folder/with/results' - -# Folder for the calculated pose errors and performance scores. -eval_path = r'/path/to/eval/folder' -``` - -with - -``` -######## Basic ######## - -# Folder with the BOP datasets. -datasets_path = str(os.environ['BOP_DATASETS_PATH']) -results_path = str(os.environ['BOP_RESULTS_PATH']) -eval_path = str(os.environ['BOP_EVAL_PATH']) -``` - - -
-This part is deprecated and will be removed - -Also, replace - -``` -# For offscreen C++ rendering: Path to the build folder of bop_renderer (github.com/thodan/bop_renderer). -bop_renderer_path = r'/path/to/bop_renderer/build' -``` - -with -``` -# For offscreen C++ rendering: Path to the build folder of bop_renderer (github.com/thodan/bop_renderer). -bop_renderer_path = /path/to/happypose/happypose/pose_estimators/megapose/deps/bop_renderer/build -``` -
- -### 2. Bop_renderer - -
-This part is deprecated and will be removed - - -This installation is tested only on Ubuntu/Debian system. Please refer to [bop_renderer](https://github.com/thodan/bop_renderer/) if needed. - -``` -cd /happypose/pose_estimators/megapose/deps/bop_renderer -sudo apt install libosmesa6-dev -conda install -c conda-forge mesalib -cmake -B build -S . -DCMAKE_BUILD_TYPE=Release -cmake --build build -``` -
- -## Usage +Please make sure you followed the steps relative to the evaluation in the main readme. This needs to be adapted (SO3_grid not used for CosyPose) diff --git a/docs/book/cosypose/test-install.md b/docs/book/cosypose/test-install.md index dcbead1c..1423718e 100644 --- a/docs/book/cosypose/test-install.md +++ b/docs/book/cosypose/test-install.md @@ -1,25 +1,6 @@ # Testing your installation -## 0. Create data directory - -``` -Create data dir /somewhere/convenient. The dataset to store are quite large. -export MEGAPOSE_DATA_DIR=/somewhere/convenient -cd $MEGAPOSE_DATA_DIR -wget https://memmo-data.laas.fr/static/examples.tar.xz -tar xf examples.tar.xz -``` - - ## 1. Download pre-trained pose estimation models - -### Megapose -Download pose estimation models to $MEGAPOSE_DATA_DIR/megapose-models: - -``` -python -m happypose.toolbox.utils.download --megapose_models -``` - -### Cosypose + ## Download pre-trained pose estimation models ```sh #ycbv @@ -34,14 +15,6 @@ python -m happypose.toolbox.utils.download --megapose_models ``` ## 2. Run examples -### Megapose - -``` -python -m happypose.pose_estimators.megapose.src.megapose.scripts.run_inference_on_example barbecue-sauce --run-inference --vis-outputs -``` - -### CosyPose - You need to download the ycbv dataset to run this example. Please see the download section. ```