March 31st, 2026 
Daniela Lopez de Luise CALL FOR PARTICIPANTS – SSBC 2026
********************************************************
10th Sclera Segmentation Benchmarking Competition 2026 (SSBC)
Foundation Models and VLMs & Label-efficient Learning
Held in conjunction with IEEE IJCB 2026
https://ijcb2026.ieee-biometrics.org/
Website: https://sites.google.com/hyderabad.bits-pilani.ac.in/ssbc2026/home
https://ijcb2026.ieee-biometrics.org/
Important dates: Registration is open now
Website: https://sites.google.com/hyderabad.bits-pilani.ac.in/ssbc2026/home
********************************************************
We are inviting all researchers and research teams to participate in our new Sclera Segmentation challenge, which focuses on Foundation Models and VLMs, as well as on label-efficient segmentation approaches.
All participants (above baseline) will coauthor the joint IJCB paper. The results submission deadline is expected by mid-May 2026.
SSBC 2026 addresses the theme of “Sclera segmentation in the age of foundation models and large multi-modal models”, as well as label-efficient learning. It is organised as a two-track competition. Participants are free to enter only a single track or compete in both. A detailed description of the two tracks is given below:
- Track 1: Foundation Models & VLMs. In this track, we aim to benchmark the performance of sclera segmentation approaches based on Foundation Models and VLMs. Participants will submit solutions that integrate large multi-modal models into biometric segmentation pipelines. We specifically encourage grounding approaches, in which textual or auxiliary prompts guide the segmentation, or novel architectures built on top of FMs/VLMs. Results will be compared against standard lighter-weight segmentation models provided by the organisers.
- Track 2: Label-Efficient Learning. The second track will focus on semi-supervised, weakly-supervised, and self-supervised learning approaches. Participants will be challenged to train models capable of learning from incomplete or no labels, addressing the critical issue of data scarcity in medical and biometric imaging. To maintain fairness during evaluation, participants may be required to submit their models and training code, which the organisers will test using a flat protocol for the segmentation task.
To enter the competition, visit the website and follow the registration instructions: https://sites.google.com/hyderabad.bits-pilani.ac.in/ssbc2026/home
Best regards on behalf of all co-organisers,
Posted in Congress