Just to clarify, I'm not currently actively working on this. The top priority for 0.8.26 is to be able to export angle tracking timelines and update the Basler module to work with Pylon 5.
692 2016-12-07 08:09:00
Re: Crash ... Logfile attached (5 replies, posted in Bug reports)
Thanks for the report, it's quite a weird error.
Try to check
C:\Users\Rohloff\AppData\Local\Kinovea\Kinovea.exe_Url_agcfeswnwwpdoul1ihhwyulj02lzin4f\0.8.15.27628\user.config
You should be able to delete user.config and even the entire AppData\Local\Kinovea directory and restart Kinovea.
This thing is created automatically by Windows and it might not be removed during uninstallation. If you try to uninstall/re-install make sure it's gone.
Unfortunately i cannot switch to a experimental Version of Kinovea
because the kamera only works in 0.8.15 (IDS U-eye)
Wait, that's new. There might have been an experimental version where I inadvertently left code to ignore the IDS cameras in the DirectShow filter while I experimented with native support, but I just checked the code and there is no "ignore" left, so the general purpose camera module should support your camera just like 0.8.15 did.
Can you locate anything about your camera in the general log.txt in 0.8.25?
693 2016-12-07 08:01:41
Re: Problems with playing two screens (3 replies, posted in Bug reports)
Hi,
Thanks for the report.
Could you make a quick test to see if version 0.8.25 has the same issue?
The auto slow down is when there is not enough resources to handle the video(s) at the nominal speed.
Did you have superposition enabled? Does it cause trouble specifically with that function enabled or all the time?
694 2016-11-28 22:21:03
Re: Experimental version - 0.8.25 (44 replies, posted in General)
Stopwatch in countdown mode within the segment prior to the synchronization point should work. However the countdown mode only works if the stopwatch has a well defined range, specified by start and end points.
I think in your case you want to do the following:
Go to the begining of the segment.
Add the stopwatch object on the first frame.
Right click the stopwatch and choose "Start"
Move forward in time up to the synchronization point (shortcut: F8).
Right click the stopwatch and choose "Stop".
Right click the stopwatch again and choose "Countdown".
695 2016-11-26 10:45:32
Re: Experimental version - 0.8.25 (44 replies, posted in General)
Thanks for the feedback!
The delay feature is alive and well and has been expanded in the latest version (see mosaic button that allow you to split the capture screen and get multiple streams from the same camera at different delay values).
What has changed is that when recording video, we previously recorded whatever was displayed in the capture screen, so if you had a 10-second delay and recorded 5 seconds of video, at the end of the recording you have on file an action that happened before you hit the record button.
The new behavior (since 0.8.24, March 2015) is to record the stream as it comes from the camera, independently of what is shown on screen. So if you have a 10-second delay and record 5 seconds of video, at the end of the recording you have the action following your hitting the record button.
This was done to improve the recording performance and avoid dropped frames. Dropped frames cause mismatches between real time and video framerate and break time and speed measurements when analyzing the videos, so it's very important to get this right.
I understand that there is now a missing feature of being able to record action in "pre-roll" fashion and considering various options.
Another feature that is often requested is to be able to increase the size of the delay buffer. This is currently limited by memory. It should be possible to increase it on 64-bit versions, but the real solution would be to allow the delay buffer to be stored on disk. I feel these two features are interlinked because if we have the delay buffer in a temporary file on disk, it should be a small step to being able to save the part of the delay buffer between the start and stop record commands to be converted to a regular video file.
696 2016-11-18 12:04:55
Re: RTSP stream (4 replies, posted in Ideas and feature requests)
It is not currently possible.
I think the way to go with this would be to have a VLC backend as capture source and leverage all the streams that VLC supports.
697 2016-11-18 11:57:44
Re: 1 camera, 2 capture screens (2 different delays) (5 replies, posted in Bug reports)
There is slow motion on a live feed indeed. It was actually the challenge that sparkled the development of this multi-view feature. In the mosaic configuration dialog the third option is Slow motion.
Obviously continuous slow motion is not sustainable per se, if you are at 0.5x speed for example, during the time you are watching the slowed down feed, twice as much data needs to be cached for later display. This can't be going on forever. The way it works is using the multiple views, each view starts at a different point in time and slowly drifts, until it jumps back to live and start drifting again. Any given action should be displayed in slow motion somewhere in one the sub-feed or maybe start in one and end in another.
Yeah record/replay scenarios would be very interesting. I'm starting to think that the way to keep it completely open to the myriad of use-cases would be to make it fully "programmable", and have a bunch of standard scripts built-in for ease of use. Then for niche use-cases people can write their own script or have someone write it for them.
698 2016-11-17 21:24:41
Re: 1 camera, 2 capture screens (2 different delays) (5 replies, posted in Bug reports)
Great. That's the type of feedback I was looking for for this feature. I put it in as an experiment with default settings but not really knowing the level of configurability that would be useful for all the use cases.
I'm not sure the other modes are that useful after all, even the slow motion one. If anyone has tried them please write up.
If not it might be better in terms of usability to focus on the multiple delays use-case and just be able to input the delay of each stream.
I feel having only two tiles instead of four will not add anything because we still need to draw them at their original aspect ratio and so they won't be able to take more space on the frame.
I also feel that supporting 25 tiles or even 16 is overkill. So maybe focusing on the 4 quadrants with custom delays would be the simplest in terms of user interface for both the user and dev since it can be hardcoded as 4 text boxes.
The underlying code would still be capable of all the other stuff but not necessarily exposed until clear user-cases are defined.
699 2016-11-16 17:54:59
Re: 1 camera, 2 capture screens (2 different delays) (5 replies, posted in Bug reports)
You may use the new feature introduced in 0.8.25 that splits the capture screen in several tiles and display the stream at different delays.
Look for the grid-like icon between the configuration and pause buttons. Select "multiple view" and choose the number of tiles (restricted to perfect squares so that it fits nicely in the original aspect ratio).
At the moment the delay of each sub stream is computed internally so that they spread equally over the memory buffer containing the delayed frames. Does your use-case involve having to manually specify the delay of each stream?
Note that it is purely a "display" feature, if you hit record it is still the original camera stream that will be recorded to disk.
700 2016-11-14 05:56:59
Re: Manual distance calibration (1 replies, posted in Ideas and feature requests)
I can think of two ways to do that:
If you add a line anywhere and do "display measure", it's going to be in pixels. So if you know your scaling factor you could compute the true size of that line and input it manually in the calibration dialog.
The second way is to open and modify the KVA file directly (You have to "save" it at least once first). Look for the node Calibration > CalibrationLine > Scale. This value is the number of real world units per pixel.
701 2016-10-04 17:15:26
Re: Experimental version - 0.8.25 (44 replies, posted in General)
I can repro this, thanks for the report. It was not intended.
702 2016-09-25 20:52:09
Re: Experimental version - 0.8.25 (44 replies, posted in General)
getpa wrote:
having installed the 0.8.25 version x64.exe, i cannot access the PS3eye camera anymore - it is not even appearing in the cam list.
With the win32 version, however, I can still access & work with the cams as previously.
Thanks for the report!
703 2016-09-25 20:49:28
Re: Experimental version - 0.8.25 (44 replies, posted in General)
TriStuff wrote:
As an Android version is unlikely, was wondering whether to buy a cheap Windows Tablet. Will this new version run on the Windows Tablet OS?
If the hardware is running Windows 10 it should work. It will not work on Windows Mobile. Some of the controls and menus might be a bit small for touch interface though.
704 2016-09-24 06:58:07
Re: Exporting Angle Tracking (6 replies, posted in General)
I confirm that 0.8.25 cannot export tracked angles data, only the static angle or individual tracked points from the trajectory tool.
705 2016-09-24 06:35:23
Re: Using 0.8.15 and 0.8.25 side by side (2 replies, posted in General)
You can use the .zip version of 0.8.25. After extraction it will operate from its directory with its own settings and program files. You can even put it on a thumb drive if you want. It's the so-called "portable application" version and doesn't need to be formally installed. Note that it should be extracted to a place where you have write rights though, so not under "C:\Program Files" for example.
In 0.8.15 there was a special export mode where the analysis data was embedded in the video as a subtitle stream. This is no longer supported directly. It should however be possible to reopen the file in 0.8.15 and re-export the analysis in a standalone file (.kva), and then import that file back into 0.8.25. I haven't tried in a while but the newer version should still have a converted to read the old data format.