At Makercon I talked about oject recognition

Here is what they have been working on. I just received the link of the github release:

https://github.com/teradeep/demo-apps

 

And of course MyRobotLab:


Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.
beetlejuice's picture

It seem to work very well !

It seem to work very well !

Alessandruino's picture

Gael, it's very impressive

Gael, it's very impressive !

the database of all the pictures they used for the training of that software is opensource?

We could train opencv haar classifiers, for object recognition if we have access to that database :)

GroG's picture

We are headed this way

We are headed this way :

Since this system as well as many others in the future will rely on remote communication - finishing the MRL API is higher up on the TODO list.

A solid API will allow all the MRL instances running in the world (potentiallly) to talk to one another

Good communication is better than fast processing... because with teamwork tasks can become significantly easier

borsaci06's picture

Looks promising....

Gael,

That is great... looks very promising.... It is very useful in a confined area such as a home or office... Not a big image database is needed... all possible images in a small working environment can be loaded to a database much easier... I will be definitely following this...

 

hairygael's picture

Yes it is very

Yes it is very promising,

Though I have no idea if the data is available as open source

GroG's picture

Look We are on a HUGE SCREEN

Look We are on a HUGE SCREEN !  

REPRESENT ! 

Nice pictures Gael - You gave an outline of what InMoov is ?
Was it well recieved?
What were the worst and best things of your trip ?