PyTorch implementation of Emotic CNN methodology to recognize emotions in images using context information.
-
Updated
Jan 8, 2024 - Python
PyTorch implementation of Emotic CNN methodology to recognize emotions in images using context information.
Code repo for the EMOTIC dataset
Project to infere emotional expressions and benchmark datasets by Niklas Wagner, Felix Mätzler, Samed R. Vossberg, Helen Schneider and Svetlana Pavlitska.
Graduation project: object detection and emotion/context categorization on the EMOTIC dataset using RetinaNet.
Reconnaître 26 émotions sur photo en décrivant aussi la scène (objets, couleurs, luminosité) quand le visage est masqué : la branche contextuelle de 114 dimensions porte le mAP à 25,60 % sur EMOTIC, soit +0,49 point.
gRPC Microservice in Python that uses a CNN model (trained with EMOTIC Dataset) to recognise emotions from persons (requires bounding box data) in images.
To associate your repository with the emotic topic, visit your repo's landing page and select "manage topics."