Commit ddfba7cb authored by Adrien Dorsaz's avatar Adrien Dorsaz

Update movie plot.

parent 9a8ef9e2
Video with Game of Thrones generic and boxplots
Game of Emotions by PocketBrain project
Who will be master of his Emotions ?
Game of Emotions
Why to become 'master of Emotions' ?
# The one who control his emotions, can...
* keep calm in any condition
* improve expression of what you want
* improve expression of how you feel
* improve how other feel you
# If you have control of your emotions, you can
* keep calm in any sitation,
* improve the expression of what you want
* understand your feelings
* and you can influence others by deciding your feelings
# So, he can be on the FER throne
Indeed, he will be "master of his emotions"
So he can take the best decision in every situation
And he'll never be manipulated by others
# So, you can sit on the iron throne
Indeed, you will be "master of your emotions"
So you can take the best decisions in every situation
And you'll never be manipulated by others
# But the question his learn to control emotions ?
# The question is now how can you learn to control your emotions ?
It can be done with help of the PocketBrain project
# With help of emotionless computers !
Indeed, The PocketBrain is a project to find emotions of people and to
sort them in different categories
# Computer to control emotions ??
## Computer can _measure_ your emotions :
* it determines how you feel with 2 values :
* arousal : are you _calm_ or _exited_ ?
* valence : feel you _positiveness_ or _negativeness_ ?
# Why "to find emotions" ?
Because we have a lot of emotions which aren't still
easy to understand self and to express to other
## But you have to train it :
* So it knows what is your _brain activity_ for each emotion you have
* And it can apply _statistics_ to know which are your current emotion
# But our body already express our emotion state with mimics and other behaviours...
Indeed, but such behaviour can be controlled and modified by people
For example, actors have to express some emotion with their body behaviour,
but they don't feel in reality this emotion
# Computer isn't really _TRANSPORTABLE_ ...
Use your Android phone !
Smartphone are _real computers_ now !
# And what about to "classify" them in categories ?
Some research have been done last years to measure emotions of people
and they find two key concept to categorize emotions :
* The arousal : are you _calm_ or _exited_ ?
* The valence : feel you _positive_ or _negative_ ?
# Wait, you've said "measure brain activity" ? We'll have to go in _hospital_ ?
No, you'll be able to do it every where with emotive headset!
# show picture of emotions around cercle on axis
# Wow, a so small headset is enough accurate ?
If you train enough your computer, you'll have some good results
# How do you find these emotion keys ?
You have to use a computer and an EEG headset :
* Computer is used to compute these keys and categorize emotions
* EEG headset is used to measure the brain electric activity
# How to translate electric activity into arousal and valence ?
The computer have to _filter_ the measures of the electric activity to have the _alpha_ and _beta_ brain waves.
Then the computer make ratio of these waves between left and right hemisphere to get arousal and valence
# Then you can directly find emotion ?
No you have to train your computer with yourself, because arousal and valence is different for everyone.
* So it has to learn what is your _brain activity_ for each emotion you want to measure
* Then, when you want to find an emotion, it can apply statistics with current measure and the learned one to find how you feel
# Let's practice ! Hum..., computers aren't movable, so it's useless.
Indeed, it's better if you can measure emotions every where...
Hey, you can use your Android phone!
Indeed, smartphone are _real computers_ now !
# Fine ! But if you search "EEG" on Google, they are intrusive and unmovalbe tools
Indeed, but Emotiv have created light EEG headset to use as gamepad
# Enough words, have someone done it ?
All on github :
All the code is on github :
sinlabsemester-2013/PocketBrain :
the application
sinlabsemester-2013/think-core :
fork of SBS2 (see below) with emokit inside
# Show me !
* capture application with Game of Thrones and explanations
* capture of PocketBrain and UI explanations : TRAIN
* capture of PocketBrain and UI explanations : TRY
## Which are key pieces ?
Which are key pieces ?
# Emocap headset
* closed software (so encrypted communication)
# Emokit open-source project
......@@ -64,6 +91,26 @@ sinlabsemester-2013/think-core
* make application for Android
# AOSP (open-source Android code)
* To use your phone as you want
# What was your roadmap ?
1. Get SBS2 and compile it
2. Try to compile hello world applications
3. Make it work with headset
4. As expected we had to use Emokit and merge its code with SBS2
5. Run hello world and the SBS2 EEG viewer
7. Try to run them on Android Emulator
Found that emulator can't reach USB devices and so code on desktop
8. Try to use the 3D Brain application
We had several bugs to run it and we've tried to fix them
But we weren't able and we asked help to main developpers but
we learned that it was a proof of concept not totally finished
9. Create our BrainPocket applications
10. Read colorOfMind documentation and apply it in our application
We had to hack SBS2 to apply two filters in same time
11. Find NaiveBayes documentation and apply it
11. Make tests and fix bugs
12. Run in real situation with a subject
# Special thanks to
* Alex Barchiesi
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment