A new competition, CircleID, has been launched for the ICDAR 2026 competition, focusing on writer identification and pen classification using only scanned hand-drawn circles. The dataset includes over 46,000 circle images from 50 known and 16 unknown writers, utilizing eight different pens. The competition featured two tasks: open-set writer identification and cross-writer pen classification, attracting hundreds of teams and thousands of submissions. The top models achieved accuracies of 64.8% for writer identification and 92.7% for pen classification, establishing a new benchmark for analyzing minimal biometric traces. AI
Summary written by gemini-2.5-flash-lite from 1 source. How we write summaries →
IMPACT Establishes a new benchmark for analyzing minimal biometric traces, potentially impacting forensic analysis and document examination.
RANK_REASON Publication of a competition paper detailing a new dataset and benchmark for writer identification and pen classification. [lever_c_demoted from research: ic=1 ai=0.7]