wellton/Emosic
1
0
Fork 0
Código Issues Pull Requests Ações Pacotes Projetos Versões Wiki Atividade

Update README.md

Ver Código-Fonte
Esse commit está contido em:
Charlie Hewitt
2018-05-08 09:07:51 +01:00
commit de GitHub
pai b17368f7fd
commit ab7590a4a6
1 arquivos alterados com 3 adições e 5 exclusões
Baixar Arquivo de Patch Baixar Arquivo de Diff Expandir todos os arquivos Colapsar todos os arquivos
README.md
+3 -5
Ver Arquivo
@@ -1,10 +1,8 @@
![Logo](/icon.png)
# Emosic
Code for the paper: [CNN-based Facial Affect Analysis on Mobile Devices](TODO).
Music recommendation app based on user affect using CNNs for emotion classification and valence/arousal regression. Song recommendations made using the [Spotify Web API](https://developer.spotify.com/web-api/).
CNNs trained on [AffectNet](http://mohammadmahoor.com/affectnet/) using [Keras](https://keras.io) and deployed to iOS by conversion to coreML using [coremltools](https://github.com/apple/coremltools).
<!--**Full write up [here]().**-->
Music recommendation app based on user affect using CNNs for emotion classification and valence/arousal regression. Song recommendations made using the [Spotify Web API](https://developer.spotify.com/web-api/). Intended as a proof-of-concept that emotionally intelligent user interfaces are now feasbile on todays high-spec mobile phones.
CNNs are trained on [AffectNet](http://mohammadmahoor.com/affectnet/) using [Keras](https://keras.io) and deployed to iOS by conversion to coreML using [coremltools](https://github.com/apple/coremltools). Networks are designed to minimise space consumption, but retain near state-of-the-art performance on AffectNet.