File size: 2,140 Bytes
eb72cfa
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
# PolyCAT Data Card

## Dataset Summary

| Property | Value |
|----------|-------|
| **Name** | PolyCAT (Polygon-Aperture Eye-Tracking Dataset) |
| **Version** | 1.0 |
| **License** | CC BY 4.0 |
| **Modality** | Eye tracking (gaze position, fixations, pupil size) |
| **Task** | Free viewing with polygon aperture constraint |
| **Size** | ~3 GB (raw) + ~12 GB (processed gaze) |
| **Participants** | 30 included |
| **Trials** | 21,026 total |
| **Stimuli** | 600 CAT2000 images x 27 polygon apertures |

## Intended Use

- Saliency prediction under geometric viewing constraints
- Gaze modeling and scanpath prediction
- Study of visual attention allocation in aperture-constrained viewing
- Benchmarking saliency models with non-rectangular viewing regions

## Data Collection

### Hardware
- EyeLink 1000+ (SR Research), head-mounted, binocular at 500 Hz per eye
- 27" 4K monitor (3840 x 2160) at 144 Hz, 70 cm viewing distance

### Participants
- 30 participants, all included
- Ages 24-29, 14 female / 16 male
- Normal or corrected-to-normal vision
- University students

### Task
Participants viewed CAT2000 images through irregular polygon apertures for 4 seconds each, with a secondary old/new memory task after each block.

### Stimuli
- 600 images from 6 CAT2000 categories: Fractal, Object, OutdoorNatural, Random, Satelite, Sketch
- 27 polygon aperture shapes with systematically varied geometric properties
- 9 fixation cue positions (3x3 grid)

## Ethical Considerations

- Study approved by institutional ethics board
- All participants provided informed consent
- Data is de-identified: no names, no identifying images, university IDs removed
- Demographic data limited to age and gender

## Known Limitations

- P05 Part B has 317/351 trials (34 trials missing due to technical issues)
- Some sessions required restart (split sessions), handled by concatenation
- Stimulus images are from CAT2000 and subject to its original licensing
- Polygon apertures may introduce edge artifacts at aperture boundaries

## Citation

See README.md for citation information.