Human and Mouse Eyes for Pupil Semantic Segmentation

A dataset composed of 11897 grayscale images of humans (4285) and mouse (7612) eyes. In different experimental conditions:  head-fixation sessions (HF: 5061), 2-photon Ca2+ imaging ( 2P: 2551), and human eyes (H: 4285). The dataset contains 1596 eye blinks, 841 images in the mouse, and 755 photos in the human datasets. Five human raters segmented the pupil in all pictures (one per image) by manual placement of an ellipse or polygon over the pupil area. Raters flagged blinks using the same code.  All the photos are illuminated using infrared (IR, 850 nm) light sources.

The dataset contains 2 folders:

'fullFrames': contains all the grayscale images in png format.

'annotation': contains a folder called 'png' with pupil mask in the red channel. There is also a file called 'annotations.csv' containing a list with a description of each file in the dataset in this folder.

Description of the fields in annotations.csv:

filename: [string] with the file name 

eye: [0,1] if true an eye is present in the picture

blink: [0,1] if true the subject is blinking

exp: [string] what kind of experiments 

w: [int] resolution width

h: [int] resolution height

roi_x: [int] roi x coordinate

roi_y: [int] roi y coordinate

roi_w: [int] roi width-height (128x128)

sub: [int] subject's label