fulyvv/pose2sim
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

a

Browse Source
This commit is contained in:
davidpagnon 2023-08-08 18:14:19 +02:00
parent 23c559f59f
commit a9733dde7c
2 changed files with 9 additions and 8 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
Pose2Sim/Demo/User/Config.toml
View File

@ -37,10 +37,10 @@ calibration_type = 'convert' # 'convert' or 'calculate'
[calibration.calculate] # coming later: 'points' [calibration.calculate] # coming later: 'points'
calculate_method = 'board' # 'board' or 'points' calculate_method = 'board' # 'board' or 'points'
[calibration.calculate.board.intrinsics] # camera properties, only needs to be done once [calibration.calculate.board.intrinsics] # camera properties, only needs to be done once
intrinsics_board_type = 'checkerboard' # 'checkerboard' ('charucoboard' not supported yet) intrinsics_board_type = 'checkerboard' # 'checkerboard' ('charucoboard' not supported yet)
overwrite_intrinsics = false # overwrite (or not) if they have already been calculated? overwrite_intrinsics = false # overwrite (or not) if they have already been calculated?
intrinsics_extension = 'avi' # any video or image extension intrinsics_extension = 'avi' # any video or image extension
extract_every_N_sec = 1 # if video, extract frames every N seconds (can be <1 ) extract_every_N_sec = 1 # if video, extract frames every N seconds (can be <1 )

13
README.md
View File

@ -12,16 +12,17 @@
# Pose2Sim # Pose2Sim
> **_News_: Version 0.4 released:** \
**Calibration used to be the main stumbling block for users, it should be easier and better now!**\
To upgrade, type `pip install pose2sim --upgrade`. You will need to update your Config.toml file.\
*N.B.:* As always, I am more than happy to welcome contributors (see [How to contribute](#how-to-contribute)).
`Pose2Sim` provides a workflow for 3D markerless kinematics, as an alternative to the more usual marker-based motion capture methods.\ `Pose2Sim` provides a workflow for 3D markerless kinematics, as an alternative to the more usual marker-based motion capture methods.\
Pose2Sim stands for "OpenPose to OpenSim", as it uses OpenPose inputs (2D keypoints coordinates obtained from multiple videos) and leads to an OpenSim result (full-body 3D joint angles). Other 2D solutions can alternatively be used as inputs. Pose2Sim stands for "OpenPose to OpenSim", as it uses OpenPose inputs (2D keypoints coordinates obtained from multiple videos) and leads to an OpenSim result (full-body 3D joint angles). Other 2D solutions can alternatively be used as inputs.
If you can only use a single camera and don't mind losing some accuracy, please consider using [Sports2D](https://github.com/davidpagnon/Sports2D). If you can only use a single camera and don't mind losing some accuracy, please consider using [Sports2D](https://github.com/davidpagnon/Sports2D).
> **__*News*__:\
Version 0.4 released:** Easier and better calibration procedure.\
To upgrade, type `pip install pose2sim --upgrade`. You will need to update your Config.toml file.\
*N.B.:* Still looking for contributors (see [How to contribute](#how-to-contribute))
<img src="Content/Pose2Sim_workflow.jpg" width="760"> <img src="Content/Pose2Sim_workflow.jpg" width="760">
@ -784,7 +785,7 @@ If you want to contribute to Pose2Sim, please follow [this guide](https://docs.g
> - [x] **Triangulation:** Evaluate which cameras were the least reliable. > - [x] **Triangulation:** Evaluate which cameras were the least reliable.
> - [x] **Triangulation:** Show which frames had to be interpolated for each point. > - [x] **Triangulation:** Show which frames had to be interpolated for each point.
> - [ ] **Triangulation:** [Undistort](https://docs.opencv.org/3.4/da/d54/group__imgproc__transform.html#ga887960ea1bde84784e7f1710a922b93c) 2D points before triangulating (and [distort](https://github.com/lambdaloop/aniposelib/blob/d03b485c4e178d7cff076e9fe1ac36837db49158/aniposelib/cameras.py#L301) them before computing reprojection error). > - [ ] **Triangulation:** [Undistort](https://docs.opencv.org/3.4/da/d54/group__imgproc__transform.html#ga887960ea1bde84784e7f1710a922b93c) 2D points before triangulating (and [distort](https://github.com/lambdaloop/aniposelib/blob/d03b485c4e178d7cff076e9fe1ac36837db49158/aniposelib/cameras.py#L301) them before computing reprojection error).
> - [ ] **Triangulation:** Multiple person kinematics (output multiple .trc coordinates files). > - [ ] **Triangulation:** Multiple person kinematics (output multiple .trc coordinates files). Triangulate all persons with reprojection error above threshold, and identify them by minimizing their displacement across frames.
> - [ ] **Triangulation:** Offer the possibility of triangulating with Sparse Bundle Adjustment (SBA), Extended Kalman Filter (EKF), Full Trajectory Estimation (FTE) (see [AcinoSet](https://github.com/African-Robotics-Unit/AcinoSet)). > - [ ] **Triangulation:** Offer the possibility of triangulating with Sparse Bundle Adjustment (SBA), Extended Kalman Filter (EKF), Full Trajectory Estimation (FTE) (see [AcinoSet](https://github.com/African-Robotics-Unit/AcinoSet)).
> - [ ] **Triangulation:** Solve limb swapping (although not really an issue with Body_25b) by using RANSAC or SDS triangulation ignoring right and left, and then decide which side points are by majority voting + giving more confidence to cameras whose plane is the most coplanar to the right/left line. > - [ ] **Triangulation:** Solve limb swapping (although not really an issue with Body_25b) by using RANSAC or SDS triangulation ignoring right and left, and then decide which side points are by majority voting + giving more confidence to cameras whose plane is the most coplanar to the right/left line.
> - [ ] **Triangulation:** Implement normalized DLT and RANSAC triangulation, Outlier rejection (sliding z-score?), as well as a [triangulation refinement step](https://doi.org/10.1109/TMM.2022.3171102). > - [ ] **Triangulation:** Implement normalized DLT and RANSAC triangulation, Outlier rejection (sliding z-score?), as well as a [triangulation refinement step](https://doi.org/10.1109/TMM.2022.3171102).