1 Topic by ChristianK

  • ChristianK
  • Member
  • Offline

Hey there experts!

I used Kinovea for the first time during some of my biomechanics units in undergrad, now I've a sports podiatrist I'm looking at using it at my current clinic to replace SiliconCoach which is 1. Expensive and 2. doesn't like to work very often.

I've tested Kinovea at the clinic and it works well with the occassional lag spike and non-responsiveness when trying to play back a just recorded video. For some of my less technologically-savvy colleagues it can be a bit daunting so my boss has got a few questions below:
1. How can I reduce the chance of lag when with patients?
2. Is it possible to do side-by-side video comparisons to look at shod vs unshod or compare different footwear and orthotics?
3. Are there any integrated capture tools for both image stills and videos that we could email off to sports teams or other practitioners effectively?

I'm fairly certain I know the answer to these questions but I'm hoping you guys could also give me some clearer explanations so I can be sure I don't miss anything.

At the moment we have a 4 camera setup and switch between the cameras to record patient's and that works really well with Kinovea though I'm not sure if the camera quality is as high.

Would it be possible to please treat me as a complete newbie and give me a bit of a rundown on how much more I can use it for and how to optimise it for a clinical setting? I'm reading through the manual as well but thought I'd post this here in case there is anyone who would also let me ask them some questions if I get stuck.

Thanks!

2 Reply by joan

  • joan
  • joan
  • Admin
  • Offline

1. Lag: there are a few axes. Using an SSD, saving the recordings to a different drive than the Windows drive, changing the recording mode in the preferences.
2. Side by side comparisons are possible yes. For example, menu View > Two playback screens.
3. Currently the capture of video is not optimized for sharing. I'm working on this for the next version, among other things. Capture images should be fine, you can save as jpeg or png and share that. If you meant capturing the whole interface like a screencast then no, you would have to use something else on top.

For multiple cameras setup there is a possible workflow where you have one instance of Kinovea for recording and another instance of Kinovea for reviewing. This can better utilize the CPU cores. Even for just one camera and one instant replay that might be another way to improve responsiveness.

Sorry I don't have more time to expand these points in too much details right now but should have more time later in the summer. Unfortunately I didn't get around to fully update the manual lately, so it's half way done in a draft version, I'll get around that at some point as well.

3 Reply by ChristianK

  • ChristianK
  • Member
  • Offline

Hi Joan,

Apologies for the delay in replying!

Good to know, I'll keep in mind SSD (keep separate to Windows drive), use two playback screens (and is it possible to have two instances of Kinovea running to show 4 views at once? I have more I'll ask a bit later on), capture images is fine, we can work with that.

Your reply has been very helpful and I was wondering if you've got any more time to expand on it if you might be able to help me out with the hardware side too or point me in the right direction for where to look.

We have been using Kinovea in our clinic for a while and it's great because it doesn't crash (though my older colleagues still don't like change and are taking some time to get used to it). I would love to know if it's possible to get Kinovea working in this way:
- Have an automated folder routing and dual-instance capture before the patient steps onto the treadmill so we have two Kinovea instances running side-by-side (E.g. one for Anterior + posterior ankle, one for sagittal + posterior global)
- Automated naming and target folders to auto-save to a dedicated local NVMe. Having a replay folder observer that can be configured so the playback screens populate automatically once a recording is finished.

Ideally it would run like this:
1. Trigger recording with hotkey or pressing the red button.
2. Capture 5-20 seconds at around 60-120 fps
3. Hitting stop immediately saves the stream and the replay folder instantly renders the files into the side-by-side playback workspaces

I would love to know also what the best cameras would be to hook up to the system as we use HDMI at the moment but I understand USB would likely be better?

Any help would be massively appreciated and you've already helped tonnes, thanks!

4 Reply by joan

  • joan
  • joan
  • Admin
  • Offline

is it possible to have two instances of Kinovea running to show 4 views at once?

Yes you can open multiple instances to have more players and cameras.

You can give names to the instances and recall them later (menu Window > Window properties…, Reopen window, Manage windows…). The general preferences are shared and synchronized between the windows, there is no "master" you can access the window manager or change the list of capture folders from any window. The concept of "workspace" is one level above that, it's for saving a whole set of windows and creating a shortcut to restart them all at once.

Ideally it would run like this:
1. Trigger recording with hotkey or pressing the red button.
2. Capture 5-20 seconds at around 60-120 fps
3. Hitting stop immediately saves the stream and the replay folder instantly renders the files into the side-by-side playback workspaces

Yes that should work. The hotkey or record button is in each window though, so if you want to start recording simultaneously in all windows it needs a different approach, easiest is via microphone trigger but there are other means.
Same for stopping the recording. The easiest is to set a predefined duration so you can't forget it and you will see a red bar reducing towards the left as a countdown.

20 second recordings is going to make large files, for your context where you'll likely have hundreds of videos it will be a concern and the NVMe will fill up. Next version will have a way to define more sensible compression settings (and improved performances to avoid frame drops). If you use USB webcam types they often have a stream that's already MJPEG encoded, this could work well for this.

The replay starts when a file is finalized in the observed folder. You have to tell it which folder to look at, it doesn't directly know about the capture windows. The linking is via the capture folder.

So in your case the first thing you want to do is define your capture folders. Then in the camera screens you tell each camera where it will save its files (using the drop down at the bottom). And for the replay screens you point them at the corresponding capture folder. Any time the capture side finalizes a recording it should be automatically loaded and played in the replay window.

Here is a tutorial that should be relevant, it was for golf but the setup is similar I think. Tutorial for golf swing automatic capture and replay.pdf

Or you could have each pair of camera + replay in its own window, if it's simpler.

For your use case there is maybe an extra layer of automation to investigate. You could have all the patients names in a CSV file, maybe exported from the main software that handles patient records. Then you import that into the "Context" (Options > Context > Import). In the capture screens you'll then get a drop down with all patient names and can specify who is being recorded. The capture folders and file name patterns can use that as variables to automatically put the files in a folder named according to the patient.

5 Reply by ChristianK

  • ChristianK
  • Member
  • Offline

Thanks so much for all your help, it's been massive!

My colleagues aren't quite as tech savvy as would be ideal and so the goalposts have shifted slightly.

The automatic replay has been great and everything's working well, now just looking at streamlining it for them.

At the moment they've been saying that occasionally after recording they won't be able to see the live feed on the capture screen side, their solution has been closing and reopening which works with the windows memory setting but I've found that also closing any extras on the capture screen also works.

With our current setup we're opted to use a switcher to swap between each view so that it all saves as one recording which makes it a bit simpler. At the moment we have it setup with one capture and one replay screen in one window.

Would there be any way to make it so that there is a single screen that shows the capture and then automatically opens the replay once you hit "stop recording?" And ideally have it so that you can just press a key or a button to start a new recording and it'll show the capture screen again?

For the NVMe we are getting IT to make a script that automatically pushes any videos older than 6 weeks into another folder and then deletes them after a certain amount of time I can't recall right now. The storage hasn't been an issue for us though, likely because our current setup only allows 30-60fps at a max of 1080p (though really it feels like 720p).

Unfortunately we're using HDMI webcams.

Apologies for if there is any confusion but I guess now we're looking at having it work such that;
1. There is a single window with one display/screen
2. We press record and it shows the live feed which we use our switcher/multi-viewer to swap between cameras to show different views (that is all setup already and working okay in one of our gait rooms but that is more of a hardware issue in the other one)
3. We hit stop record and in the screen that was showing the live feed it now shows the live replay which we can slow down, annotate, etc
4. We have some way of starting a recording again and it goes back to the live feed screen and we no longer see the replay screen (this would also help for patient privacy)

The golf option would have been great if I could trust my colleagues to navigate more than one window haha. I might've overestimated their technological literacy...

Thank you for being so patient!

6 Reply by joan

  • joan
  • joan
  • Admin
  • Offline

It's valid, this request has come up in the past as well. I realize the current system is a bit oriented towards the more advanced use-case at the cost of making the simpler case more complicated, and that's something I'm not happy about, it breaks the rule.

To confirm, right now it's not possible to have a single-screen window that would automatically switch between capture and replay within itself.

This needs a way to hide one of the screens while we work on the other. There is a similar need in the context of comparing two videos with blending, right now it's just showing the blended view in both screens which is very redundant. It would be better to just show the left screen, and be able to unhide the other one for any operation specific to it.

So it's something I am considering, but unlikely for the next release unless I can find a low stakes change supporting it.
I think any additional buttons or menu to do this would clash in meaning with the existing primary toolbar at the top and the View menu to configure screens. When someone wants to switch back to the capture side for example they will likely want to click the button that shows one camera, but currently this will fully close the player if there is one.
So my feeling is that the solution is to change the meaning of this primary toolbar to be show/hide rather add/close.