Closed splasky closed 3 years ago
Firstly @splasky just what to say this would be a awesome feature!
There maybe some misunderstandings related to the video upload.
In order to upload usable data using the video format one needs to write the metadata file. The backend only can make sense of a sequence containing videos if it has metadata file.
Creating metadata file can be much easier achieved using the new metadata parser classes that were just merged into master.
Firstly you should make a dedicated parser class from video to sensor data instances then add the sensors data instances to a metadata parser instance and call the serialize method on that metadata parser. Uploading the generated metadata file will make the sequence usable by the backend.
This would be awesome! If you need help i am more than happy to help.
I am really curious to see an example of video that contains GPS data or other sensor data. Can you share a short example?
The mp4 videos needs track info before uploading.
Before uploading the mp4 video record to openstreetcam, the mp4 file needed to extract the GPS information.
This commit add the exiftool package to extract GPS information from mp4 file and implemented the videodiscover for uploading.