(Official Website: EMOTIC)
Watch a ~3 minute video (just me speaking, no presentations or anything) summarizing my thesis on this topic.
- YouTube Link: Read Thesis in 4-Minutes
-
Dataset download link: PAMI_19 Dataset
-
Baseline CNN model : PAMI_19 MODEL, where you will also find a pseudo code for testing your images.
-
Annotations download link: PAMI_19 Annotations
-
Dataset download link: CVPR_17 Dataset
-
Baseline CNN model: CVPR_17 MODEL, where you will also find a pseudo code for testing your images.
-
Annotations download link: CVPR_17 Annotations
-
Demo for computing the Average Precision (AP) demoComputeAP
-
Single Image Inference for PAMI version (input image and the corresponding bounding box info required): Single Image Inference
-
For Model architecture (same for cvpr and pami publications), you can refer emotic_cnn_model_structure.txt
-
PAMI'19 paper: Context Based Emotion Recognition using EMOTIC Dataset
-
CVPR'17 paper: Emotion Recognition in Context
-
CVPRW'17 paper: EMOTIC: Emotions in Context Dataset
-
PhD Manuscript: Visual Scene Context in Emotion Perception [much easier to read with broader set of experiments]