AI search engine for video footage

[Brian Timmons]

Jumper - Jump to the best parts of editing.

Say goodbye to hours wasted on naming, tagging, sorting, and endless searches for footage. Jumper helps you find and edit the best parts instantly.

getjumper.io

Workflow extension that allows one to use AI to search footage for specified
image and audio content (i.e sound bites).

Might not sound exciting but if you edit projects with several hours of footage you
quickly see the value of this type of tool.

This is the type of AI product I'm a fan of.

Haven't tried it yet but look forward to using it.

I have no affiliation with the people that make this. Saw the announcement on another site and spreading the news
as I think this can truly help people's editing.

Brian Timmons
BRITIM/MEDIA

 

[rand thompson]

In there Youtube comments, They say that Jumper is still in development for Davinci Resolve and Avid Media Composer.

 

[Brian Timmons]

In there Youtube comments, They say that Jumper is still in development for Davinci Resolve and Avid Media Composer.

Yes- the site says it's in Private Beta.
I think it might be because Jumper works as a workflow extension (FCP and Premiere support) instead of a traditional plug-in.
I'm not aware of Avid and Resolve having workflow extension support so they are likely creating a workaround, hopefully with no compromises.

edit: Resolve does have a workflow integration so perhaps not the above.

Brian Timmons
BRITIM/MEDIA

 

[rand thompson]

In the comments below they say those are still in development.

 

[Brian Timmons]

good

 

[MaxLund]

Hi everyone, I'm the lead developer of Jumper. I just wanted to let you know we recently added support for .R3D files, given that you have REDCINE-X installed on your system. You can try it out for free at https://getjumper.io

You have our documentation pages and a longer tutorial video here https://docs.getjumper.io/videos/
If you have any questions or need help with anything you can join our Discord here https://discord.gg/3JFNYAfwSb or send an email to hello@getjumper.io

 

[Patrick Tresch]

Thanks for the link! I wonder how does it analyse the footage. Does it need to be online and using a proxy database?

 

[MaxLund]

Thanks for the link! I wonder how does it analyse the footage. Does it need to be online and using a proxy database?

Nope, it works completely locally on your device - which works out fine since you editors usually have pretty beastly machines

You only need to be online to validate your license keys. Free trial licenses need to keep being online since we keep track of how many searches you have made (100 free searches included). But for paid licenses you can literally validate your license key once and then shut off your internet, and Jumper will keep working as usual.

 

[Patrick Tresch]

Impressive. Playing around with FCP waiting for Resolve ;-)

 

[Patrick Tresch]

Would it work with an audio library that is stored on a local drive? Or would it have to be linked to an audio editing tool (Pyramix, Protools... ) ?

 

[MaxLund]

Impressive. Playing around with FCP waiting for Resolve ;-)

Going to try to make a push to get that out early next year.

Would it work with an audio library that is stored on a local drive? Or would it have to be linked to an audio editing tool (Pyramix, Protools... ) ?

We support most of the usual video and audio file formats for the speech-to-text processing and transcription search. For the visual processing and search we support the same video file formats, as well as many common still image file formats. Not sure if that answers your question?

The easiest way to find out if something works is just to get a free trial license at getjumper.io and test it. The trial version
has all features included - it's just limited to 100 searches.

 

[Patrick Tresch]

I was wondering if it could find : wind in the tree, sea gull, city hum, waves braking, dog barking, children playing, football match... such type of sounds?

 

[MaxLund]

I was wondering if it could find : wind in the tree, sea gull, city hum, waves braking, dog barking, children playing, football match... such type of sounds?

No, the transcription searches only works for spoken words and music (found by searching "music"). However the visual search should be able to find all of the examples you listed, given that the visual content in the footage matches what sounds is being heard in that moment.