Skip to main content
Archived

Support Full Motion Video

  • September 11, 2017
  • 5 replies
  • 67 views

fmelizard
Safer
Forum|alt.badge.img+22

It might be interesting if FME could support Full Motion Video.

There are a number of ways that FME could be involved, such as: clip by time, clip by location, sample / extract frames as rasters, combine, split, resize, georeference, extract georeferencing, embed georeferencing, sync video and audio.

Then there may be standards emerging that FME could utilize.

How would you like to use FME with Full Motion Video?

This post is closed to further activity.
It may be an old question, an answered question, an implemented idea, or a notification-only post.
Please check post dates before relying on any information in a question or answer.
For follow-up or related questions, please post a new question or idea.
If there is a genuine update to be made, please contact us and request that the post is reopened.

5 replies

ulfme
Contributor
Forum|alt.badge.img+15
  • Contributor
  • September 13, 2017

This is a really useful thing. (Have been doing that with using FFMPEG with FME for a couple of years. There are so many usecases). But in short - any automation of video editing is welcome. Regular video-editing is a very manual and time-consuming task and there is tons of unused videos out there.


fmelizard
Safer
Forum|alt.badge.img+22
  • Author
  • Safer
  • October 13, 2017
Some refer to this as geotagged video. In case they are searching for this idea.

 

 


gav
Contributor
Forum|alt.badge.img+10
  • Contributor
  • March 27, 2018

In my spare time, I've captured video with a Garmin Virb and am using the same technique ulfme talks about and is detailed here: https://blog.safe.com/2017/11/fme-spatially-enabled-video-editor/

Personally, I am working on mapping the trees in my neighborhood. Right now, I have the images georeferenced to the track, and manually marked the base of the tree with a magenta raster cell.

Any transformers to determine distance using the adjacent track line(s) bearing, angle of the camera, and raster cell mark would be super-helpful. If a raster line (in a distinct color) could be drawn across the tree to estimate age, that would be a 10x bonus!


bwhite
Contributor
Forum|alt.badge.img+2
  • Contributor
  • August 31, 2021

I'd like to add a few suggestions if I may. I worked at a company that received various large uncompressed video formats for CCTV footage taken for performing engineering assessments from within sewer pipes. But these ideas would be useful for other purposes.

This is a wish list, but it would be good to have transformers that could:

  • Compressing and converting the footage from multiple compressed formats to a format more suitable for storage on a network.
  • Extract every nth frame. The aforementioned uncompressed footage was viewed as a motion video when an image format, taken every X frames, or every Y metres (sometimes the CCTV video was measured in distance) would be more suitable for engineering assessments.
  • Assign spatial links from GIS data to the CCTV footage. For instance, a spatial dataset recording the position and direction of the camera.
  • Extract image properties, file name, path etc. to be written to a database to create hyperlinks to the footage.
  • extract, remove, realign, or assign audio in a movie file. For example, an engineer might remove water flow noise from a video, and then add a commentary track describing any repairs needed.

Thanks


  • September 28, 2022

Totally agree with you, guys! That could be a really useful feature. I’m happy to be not the only one considering this feature extremely helpful. Actually, I totally agree with ulfme’s thoughts on video automation. I mean, everything that could speed up the working process is very welcome, actually. For example, a few days ago, I finally found a vhs to digital converter, so now it will take me twice less time to do my work. So, let’s keep this thread updated, guys. Best wishes to you all.