Kanizsa Illusory Contour (kani2) Dataset with Border-Ownership Annotations

#  Annotations for an illusory Kanizsa (kani2) dataset

• This repository contains annotations for an illusory Kanizsa (kani2) dataset. These annotations were developed to
    support research in **border ownership representation**, **disparity-based segmentation**, and **figure-ground
    organization** in synthetic depth stimuli.  
• The dataset was used in both Chen et al. (SciRep, 2025) and Chen et al. (bioRxiv, 2025)  

> NOTE: This dataset complements the previously released  
> Chen T. (2025). *Border Ownership and Category Annotation Extended from Virtual KITTI 2* [Data set],  
> Zenodo DOI: [https://doi.org/10.5281/zenodo.15375453](https://doi.org/10.5281/zenodo.15375453)   
> Chen, T. (2025). *Random Dot Stereogram (rds4) Dataset with Border Ownership Annotation (Version v1)* [Data set].
> Zenodo. DOI: https://doi.org/10.5281/zenodo.16423201 

---

##  Contents

This dataset provides the following additions:
- **Border ownership maps** — 2-channel directional annotations representing ownership along contours (Chen et al., 2022)
- **Disparity maps** — Absolute (Near) disparity representations (synthetic disparity=2 pixels)  
- **Instance-level contour maps** — Extended labeling for object-level border continuity
- **Scripts** — MATLAB and Python utilities to read, write and display `.flo`-formatted annotations

> For a complete specification of annotation file formats, see [DATAFORMAT.md](./DATAFORMAT.md).

---

##  License

This dataset is released under the **Creative Commons Attribution-ShareAlike 4.0 International (CC-BY-SA 4.0)** license.

By using this dataset, you agree to:
- Attribute this dataset appropriately
- Share any derivative work under the **same ShareAlike terms (CC-BY-SA 4.0)**

See `LICENSE.txt` for full legal terms.

---

##  References

- Chen, T., Cheng, X., & Tsao, T. (2022). *Border Ownership, Category Selectivity and Beyond*.  
  In: Bebis, G. et al. (eds), Advances in Visual Computing. ISVC 2022, LNCS, vol 13599, Springer, Cham.  
  [https://doi.org/10.1007/978-3-031-20716-7_3](https://doi.org/10.1007/978-3-031-20716-7_3)

- Tianlong Chen, Xuemei Cheng, Thomas Tsao et al. *Understanding and Simulating Border Ownership Centered Segmentation*,
   Sci Rep 15, 43881 (2025). https://doi.org/10.1038/s41598-025-18843-9

- Tianlong Chen and Xuemei Cheng (2025). *A Unified Mechanism for Depth Perception and Figure-Ground Organization*,
  bioRxiv preprint (2025). https://doi.org/10.64898/2025.12.10.693427  

---

##  Contact

For questions or collaboration inquiries, please contact:

**Tianlong Chen**  
Email: [tc@tichen.net](mailto:tc@tichen.net)