The following diffs were needed to make it work under Linux

+++ b/freemocap/webcam/camsetup.py
@@ -21,7 +21,7 @@ class VideoSetup(threading.Thread):
         camName = "Camera" + str(self.camID)
 
         cv2.namedWindow(camName)
-        cap = cv2.VideoCapture(self.camID, cv2.CAP_DSHOW)
+        cap = cv2.VideoCapture(self.camID, cv2.CAP_ANY)
         cap.set(cv2.CAP_PROP_FRAME_WIDTH, resWidth)
         cap.set(cv2.CAP_PROP_FRAME_HEIGHT, resHeight)
         cap.set(cv2.CAP_PROP_EXPOSURE, exposure)
diff --git a/freemocap/webcam/checkcams.py b/freemocap/webcam/checkcams.py
index dda1348..fb4a6a0 100644
--- a/freemocap/webcam/checkcams.py
+++ b/freemocap/webcam/checkcams.py
@@ -3,7 +3,7 @@ import cv2
 
 
 def TestDevice(source):
-    cap = cv2.VideoCapture(source, cv2.CAP_DSHOW)
+    cap = cv2.VideoCapture(source, cv2.CAP_ANY)
     # if cap is None or not cap.isOpened():
     # print('Warning: unable to open video source: ', source)
 
diff --git a/freemocap/webcam/startcamrecording.py b/freemocap/webcam/startcamrecording.py
index a0cdec1..011d853 100644
--- a/freemocap/webcam/startcamrecording.py
+++ b/freemocap/webcam/startcamrecording.py
@@ -44,7 +44,7 @@ def CamRecording(
     flag = False
 
     cv2.namedWindow(camID)  # name the preview window for the camera its showing
-    cam = cv2.VideoCapture(camInput, cv2.CAP_DSHOW)  # create the video capture object
+    cam = cv2.VideoCapture(camInput, cv2.CAP_ANY)  # create the video capture object
     # if not cam.isOpened():
     #         raise RuntimeError('No camera found at input '+ str(camID))
     # pulling out all the dictionary paramters

But while everything works with one camera at a time, it seems like it blocks with two identical cameras

The error manifests itself outside of your code

I get

[  806.191512] uvcvideo: Failed to query (SET_CUR) UVC control 4 on unit 1: -32 (exp. 4).

Here is lsusb

[email protected]:~/Desktop$ lsusb 
Bus 001 Device 005: ID 046d:09a4 Logitech, Inc. QuickCam E 3500
Bus 001 Device 004: ID 046d:09a4 Logitech, Inc. QuickCam E 3500
Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
Bus 004 Device 002: ID 046d:c542 Logitech, Inc. Wireless Receiver
Bus 004 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
Bus 003 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
Bus 002 Device 004: ID 060b:7a16 Solid Year MD800
Bus 002 Device 003: ID 0627:0001 Adomax Technology Co., Ltd QEMU USB Hub
Bus 002 Device 002: ID 0409:55aa NEC Corp. Hub
Bus 002 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub

do each of the cameras need to be on a separate bus?

My team and I are attempting to use FreeMoCap for a system that records eight views in synchrony. We are currently having issues calibrating all eight cameras, which face opposing directions in our hallway set-up, making it difficult for them to see one CharUco board at once. To resolve this, we printed a double-sided CharUco board, allowing us to create a 3D skeleton output; however, the skeleton seems to be both inverted and translated compared to the subject from certain camera angles. We then tried to calibrate our system using hallway cameras on one side, which all face each other and can observe a charUco board simultaneously. Still, we saw the output skeleton as translated and inverted, ruling out the double-sided board as the reason for the translation. Any suggestions on how to calibrate a system set up like this or why the skeletons seem to be inverted/displaced?

Attached is a link with the videos and data array/calibration files for the 8 camera set-up. https://drive.google.com/drive/folders/1wv4RHzPFDLeIXr9xONC72fhbW3MKQT_q?usp=sharing

I started a new FMC session, yet when running Stage 5 FMC keeps using the .blend from my original project instead of the new one i made for it in the right session folder. It then fails saying that the file doesn't exist. Any thoughts? *replaced login with X

Running sesh 2 from /home/X/bac_mj_ar/FreeMocap_Data ──────────────────────────────────────────────────────────────────────────────── Using blender executable located at: /home/X/bac_mj_ar/FreeMocap_Data/sesh/Session_ID.blend Skipping Video Recording Skipping Video Syncing Skipping Calibration Skipping 2d point tracking ──────────────────────────────────────────────────────────────────────────────── ────────────────────────────── EXPORTING FILES... ────────────────────────────── ─ Hijacking Blender's file format converters to export FreeMoCap data as vari… ─ ──────────────────────────────────────────────────────────────────────────────── ╭───────────────────── Traceback (most recent call last) ──────────────────────╮ │ /home/X/.conda/envs/freemocap-env/lib/python3.7/site-packages/freemoca │ │ p/fmc_runme.py:335 in RunMe │ │ │ │ 332 │ │ │ │ │ │ │ │ │ │ command_str, │ │ 333 │ │ │ │ │ │ │ │ │ │ shell=False, │ │ 334 │ │ │ │ │ │ │ │ │ │ stdout=subprocess.PIPE, │ │ ❱ 335 │ │ │ │ │ │ │ │ │ │ stderr=subprocess.PIPE │ │ 336 │ │ │ │ │ │ │ │ │ │ ) │ │ 337 │ │ │ │ while True: │ │ 338 │ │ │ │ │ output = blender_process.stdout.readline() │ │ │ │ /home/X/.conda/envs/freemocap-env/lib/python3.7/subprocess.py:800 in │ │ init │ │ │ │ 797 │ │ │ │ │ │ │ │ p2cread, p2cwrite, │ │ 798 │ │ │ │ │ │ │ │ c2pread, c2pwrite, │ │ 799 │ │ │ │ │ │ │ │ errread, errwrite, │ │ ❱ 800 │ │ │ │ │ │ │ │ restore_signals, start_new_session) │ │ 801 │ │ except: │ │ 802 │ │ │ # Cleanup if the child failed starting. │ │ 803 │ │ │ for f in filter(None, (self.stdin, self.stdout, self.stde │ │ │ │ /home/X/.conda/envs/freemocap-env/lib/python3.7/subprocess.py:1551 in │ │ _execute_child │ │ │ │ 1548 │ │ │ │ │ │ err_msg = os.strerror(errno_num) │ │ 1549 │ │ │ │ │ │ if errno_num == errno.ENOENT: │ │ 1550 │ │ │ │ │ │ │ err_msg += ': ' + repr(err_filename) │ │ ❱ 1551 │ │ │ │ │ raise child_exception_type(errno_num, err_msg, er │ │ 1552 │ │ │ │ raise child_exception_type(err_msg) │ │ 1553 │ │ 1554 │ ╰──────────────────────────────────────────────────────────────────────────────╯ FileNotFoundError: [Errno 2] No such file or directory: '/home/X/bac_mj_ar/FreeMocap_Data/sesh/Session_ID.blend --background --python /home/X/.conda/envs/freemocap-env/lib/python3.7/site-packages/fre emocap/freemocap_blender_megascript.py -- /home/X/bac_mj_ar/FreeMocap_Data/sesh 2 0': '/home/X/bac_mj_ar/FreeMocap_Data/sesh/Session_ID.blend --background --python /home/X/.conda/envs/freemocap-env/lib/python3.7/site-packages/fre emocap/freemocap_blender_megascript.py -- /home/X/bac_mj_ar/FreeMocap_Data/sesh 2 0'

Seems like the threading is the cause of the issue here.

(freemocap) MacBook-Pro:freemocap sam$ python runme_FreeMoCap.py 
Starting initialization for stage 1
  0%|                                                    | 0/20 [00:00<?, ?it/s]Oct  2 15:40:31  ThetaUVC_blender[33664] <Notice>: ------------ ThetaUVC_blender plugin start (version:2.0.1 built:Fri Aug 19 15:54:46 JST 2016 pid=33664 RELEASE). ------------ #thetauvc
  5%|██▏                                         | 1/20 [00:01<00:37,  1.98s/it]OpenCV: out device of bound (0-0): 1
OpenCV: camera failed to properly initialize!
OpenCV: out device of bound (0-0): 2
OpenCV: camera failed to properly initialize!
OpenCV: out device of bound (0-0): 3
OpenCV: camera failed to properly initialize!
OpenCV: out device of bound (0-0): 4
OpenCV: camera failed to properly initialize!
OpenCV: out device of bound (0-0): 5
OpenCV: camera failed to properly initialize!
OpenCV: out device of bound (0-0): 6
OpenCV: camera failed to properly initialize!
OpenCV: out device of bound (0-0): 7
OpenCV: camera failed to properly initialize!
OpenCV: out device of bound (0-0): 8
OpenCV: camera failed to properly initialize!
OpenCV: out device of bound (0-0): 9
OpenCV: camera failed to properly initialize!
OpenCV: out device of bound (0-0): 10
OpenCV: camera failed to properly initialize!
 55%|███████████████████████▋                   | 11/20 [00:02<00:01,  7.18it/s]OpenCV: out device of bound (0-0): 11
OpenCV: camera failed to properly initialize!
OpenCV: out device of bound (0-0): 12
OpenCV: camera failed to properly initialize!
OpenCV: out device of bound (0-0): 13
OpenCV: camera failed to properly initialize!
OpenCV: out device of bound (0-0): 14
OpenCV: camera failed to properly initialize!
OpenCV: out device of bound (0-0): 15
OpenCV: camera failed to properly initialize!
OpenCV: out device of bound (0-0): 16
OpenCV: camera failed to properly initialize!
OpenCV: out device of bound (0-0): 17
OpenCV: camera failed to properly initialize!
OpenCV: out device of bound (0-0): 18
OpenCV: camera failed to properly initialize!
OpenCV: out device of bound (0-0): 19
OpenCV: camera failed to properly initialize!
100%|███████████████████████████████████████████| 20/20 [00:02<00:00,  9.17it/s]
2021-10-02 15:40:41.571 python[33664:261625353] WARNING: NSWindow drag regions should only be invalidated on the Main Thread! This will throw an exception in the future. Called from (
	0   AppKit                              0x00007fff22b6ded1 -[NSWindow(NSWindow_Theme) _postWindowNeedsToResetDragMarginsUnlessPostingDisabled] + 352
	1   AppKit                              0x00007fff22b58aa2 -[NSWindow _initContent:styleMask:backing:defer:contentView:] + 1296
	2   AppKit                              0x00007fff22b5858b -[NSWindow initWithContentRect:styleMask:backing:defer:] + 42
	3   AppKit                              0x00007fff22e6283c -[NSWindow initWithContentRect:styleMask:backing:defer:screen:] + 52
	4   cv2.cpython-37m-darwin.so           0x000000010ead0c94 cvNamedWindow + 564
	5   cv2.cpython-37m-darwin.so           0x000000010eacdd1a _ZN2cv11namedWindowERKNS_6StringEi + 58
	6   cv2.cpython-37m-darwin.so           0x000000010dc4a487 _ZL23pyopencv_cv_namedWindowP7_objectS0_S0_ + 231
	7   python                              0x0000000104cd3f2b _PyMethodDef_RawFastCallKeywords + 395
	8   python                              0x0000000104e0b9bb call_function + 251
	9   python                              0x0000000104e034eb _PyEval_EvalFrameDefault + 20171
	10  python                              0x0000000104cd3b15 _PyFunction_FastCallKeywords + 229
	11  python                              0x0000000104e0b977 call_function + 183
	12  python                              0x0000000104e03462 _PyEval_EvalFrameDefault + 20034
	13  python                              0x0000000104cd3b15 _PyFunction_FastCallKeywords + 229
	14  python                              0x0000000104e0b977 call_function + 183
	15  python                              0x0000000104e03462 _PyEval_EvalFrameDefault + 20034
	16  python                              0x0000000104cd3b15 _PyFunction_FastCallKeywords + 229
	17  python                              0x0000000104e0b977 call_function + 183
	18  python                              0x0000000104e03462 _PyEval_EvalFrameDefault + 20034
	19  python                              0x0000000104cd1fea _PyFunction_FastCallDict + 234
	20  python                              0x0000000104cd66ba method_call + 122
	21  python                              0x0000000104cd445f PyObject_Call + 127
	22  python                              0x0000000104ef7e3a t_bootstrap + 122
	23  python                              0x0000000104e7c764 pythread_wrapper + 36
	24  libsystem_pthread.dylib             0x00007fff2031a8fc _pthread_start + 224
	25  libsystem_pthread.dylib             0x00007fff20316443 thread_start + 15
)
__________________________________________
cv2::videocapture properties for Camera# 0
CV_CAP_PROP_FRAME_WIDTH: '1280.0'
CV_CAP_PROP_FRAME_HEIGHT : '720.0'
CAP_PROP_FPS : '30.0'
CAP_PROP_EXPOSURE : '0.0'
CAP_PROP_POS_MSEC : '0.0'
CAP_PROP_FRAME_COUNT  : '0.0'
CAP_PROP_BRIGHTNESS : '0.0'
CAP_PROP_CONTRAST : '0.0'
CAP_PROP_SATURATION : '0.0'
CAP_PROP_HUE : '0.0'
CAP_PROP_GAIN  : '0.0'
CAP_PROP_CONVERT_RGB : '0.0'
__________________________________________
2021-10-02 15:40:42.898 python[33664:261625353] WARNING: nextEventMatchingMask should only be called from the Main Thread! This will throw an exception in the future.

I have some undergrad students using this for a class project, and they ran into two installation issues that were easily solved, and should be solvable by either modifying the installation requirements, or just editing the instructions.

All on windows.

The first is that ipython is not automatically installed, so the start instructions fail. For those with new conda installations, after activating the env, conda install ipython is a simple fix.

The second error comes after recording, with the end of the Traceback being:

~\Miniconda3\envs\freemocap\lib\site-packages\moviepy\video\io\ffmpeg_writer.py in __init__(self, filename, size, fps, codec, audiofile, preset, bitrate, withmask, logfile, threads, ffmpeg_params)
     86             '-s', '%dx%d' % (size[0], size[1]),
     87             '-pix_fmt', 'rgba' if withmask else 'rgb24',
---> 88             '-r', '%.02f' % fps,
     89             '-an', '-i', '-'
     90         ]

TypeError: must be real number, not NoneType

This is a problem with moviepy not finding ffmpeg. It's possible to install ffmpeg on windows and add to the path independently, but I prefer to install it in conda with conda install ffmpeg. However, moviepy can't find it, so after installing ffmpeg, running

pip uninstall moviepy
pip install moviepy

works. I didn't have time to try to install ffmpeg first, but I think just adding it to the env creation line should fix the problem.

Hello,

Could someone help me solve this problem, please! Thank you in advance for your help.

Here is what displays to me, knowing that I used two webcam type cameras.

Hello! We are running into an issue in which FMC seems to be swapping the skeletons generated from one camera onto the view of another camera (see video). Is this a potential result of using greyscale cameras? The calibration seems to work fine as the video generated reflects accurate skeletons, just placed on the wrong camera (ie camera 1 generates skeleton 1, but skeleton 1 is placed onto the view of camera 3). Has anyone encountered similar issues/have any suggestions on how to resolve this issue? Thank you in advance!

https://user-images.githubusercontent.com/114196168/201419033-18be83ad-e852-4a82-b116-9fbfe470bec2.mp4

Currently, if no charuco points are detected, the code fails in this way -

Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "C:\Users\jonma\Dropbox\GitKrakenRepos\freemocap\freemocap\__init__.py", line 124, in RunMe
    sesh, board
  File "C:\Users\jonma\Dropbox\GitKrakenRepos\freemocap\freemocap\calibrate.py", line 84, in CalibrateCaptureVolume
    error,merged = cgroup.calibrate_videos(vidnames, board)
  File "C:\Users\jonma\Dropbox\GitKrakenRepos\freemocap\freemocap\fmc_anipose.py", line 1740, in calibrate_videos
    **kwargs
  File "C:\Users\jonma\Dropbox\GitKrakenRepos\freemocap\freemocap\fmc_anipose.py", line 1672, in calibrate_rows
    objp, imgp = zip(*mixed)
ValueError: not enough values to unpack (expected 2, got 0)

...which should instead -

• [ ] Raise an informative Error/Exception/Whatever thing
• [ ] (Optional) If only one camera is selected, this is a warning. If more than one selected, this is an Error
• [ ] (Optional) maybe in general there should be a 'single-camera' mode? so that people can just use this wrapper on their single webcam to play with OpenPose, MediaPipe, DLC, etc?

Hi, thanks for all your work on this project. I'm excited to see where it goes.

I'm having an issue where I can't seem to get this working with multiple cameras. I have three webcams plugged directly into my laptop (I also tried plugging into a hub, but didn't change anything).

When I go through the camera setup process and click "Submit", only the first camera lights up and only a blank rectangular window appears.

This is the output I see:

__________________________________________
cv2::videocapture properties for Camera# 0
CV_CAP_PROP_FRAME_WIDTH: '640.0'
CV_CAP_PROP_FRAME_HEIGHT : '480.0'
CAP_PROP_FPS : '30.0'
CAP_PROP_EXPOSURE : '0.008200820082008202'
CAP_PROP_POS_MSEC : '0.0'
CAP_PROP_FRAME_COUNT  : '-1.0'
CAP_PROP_BRIGHTNESS : '0.5019607843137255'
CAP_PROP_CONTRAST : '0.12549019607843137'
CAP_PROP_SATURATION : '0.12549019607843137'
CAP_PROP_HUE : '-1.0'
CAP_PROP_GAIN  : '0.20392156862745098'
CAP_PROP_CONVERT_RGB : '1.0'
__________________________________________

It looks like it's getting stuck on the first camera and never manages to start up the other cameras?

It works fine if I select just one camera.

Workflow:

1. Use mkdocs to create our knowledge base skin
2. Use HackMd with MkDocs configuration protocols to write KB articles

Notes: https://www.mkdocs.org/getting-started/

PR #212

Hey, I've followed the setup but I'm having the following on Stage 3 - Calibrate Capture Volume: list index out of range.

Here are some screenshots of the error: .

Thanks a lot for any help.

Following the process in the documentation for processing previously recorded videos, videos with the file extension MP4 are not recognized on Mac/Linux. The non-capitalized variant mp4 is recognized by Mac/Linux, and both cases should be recognized by windows (although I can't test this personally). The case sensitivity issue is due to the changing behavior of glob.glob() across operating systems, as explained here.

To resolve this issue, the file search should either be made case insensitive on Mac/Linux to match the Windows behavior, or case sensitive on Windows to match the Mac/Linux behavior (the linked article above describes a function to make glob case sensitive on Windows).

Have successfully gotten FMC off the ground for our project, but we had some questions regarding the output files produced.

1. In the Mediapipe_body+3d+xyz.csv file produced via the Alpha GUI, what is the unit for time? We see that our videos are roughly 19 seconds long and contain 2063 frames, yet we have 1753 measurements of each key point. Is this the number when all cameras are active and tracking?
2. How are the x,y,z coordinates established? Are they consistent between videos using the same calibration or different calibrations within the same hallway?

EDIT: For the Mediapipe_body+3d+xyz.csv, also wondering what the coordinates represent/what the unit for each measurement is.

Life improvement:

What? Make an option to start and stop a recording after a set duration.

e.g. pres button, recording starts after 5 seconds and ends 30 seconds after that.

Why? Because it makes recording alone easier (pres button, walk to recording volume, record) Because it can help standardize patient recordings (e.g. 30 seconds sit-to-stand test)

BONUS POINTS if FreeMoCap plays a sound when the recordings start and stop.

Down the line, users should be able to control the start and orientation of the global coordinate system using the Charuco board (https://github.com/freemocap/freemocap/issues/282).

Until that is implemented, a workable fix could be to allow users to select what camera is being used for the origin and orientation of the global coordinate system.

This will allow users to have some control over the global coordinate system by having one camera set at a specific height and with a level orientation.

It's not perfect, but it would be a great improvement.

Add a final option during calibration, "Set Floor"

Place the charuco board on the floor and use the middle (or a set corner) as 0,0,0 in the global coordinate system. Use the orientation of the charuco board to assign x and z axis (y - axis will straight up from the board).

This would greatly improve FreeMoCap's use in biomechanical settings.

Adding a sample of the blender output somewhere in the repository (or in the documentation...) would be great for helping folks see what freemocap produces!