CK+ Dataset

The Files has been resized to 48×48 pixels, converted to grayscale, and cropped to focus on faces using the haarcascade_frontalface_default method.

Emotions are defined as determined index below

• 0: Anger (45 samples)
• 1: Disgust (59 samples)
• 2: Fear (25 samples)
• 3: Happiness (69 samples)
• 4: Sadness (28 samples)
• 5: Surprise (83 samples)
• 6: Neutral (593 samples)
• 7: Contempt (18 samples)

Pixels contains 2304 pixels (48×48) each row.

Usage is determined as Training(80%) / PublicTest(10%) / PrivateTest(10%)

Any problem with dataset feel free to ask.
Updates will only happen to improve data, as collaborators show up to help.

Conclusion

Globose Technology Solutions Private Limited acknowledges the significant role of the  advancing facial expression recognition. By utilizing CK+ data, we improve the accuracy of our facial recognition systems, providing state-of-the-art technology for emotion analysis across various industries. Committed to innovation, we remain at the forefront of dataset developments like CK+, ensuring cutting-edge solutions for our clients.

Download the dataset now!