The 3DTeethLand challenge, held at MICCAI 2024, aimed to advance deep learning techniques for detecting dental landmarks from 3D intraoral scans. This challenge provided a new dataset of 340 scans to benchmark algorithms for this crucial task in orthodontics. Forty-nine teams competed, with the top-ranked team achieving a score of 0.91 using a novel two-stage Transformer approach. AI
Summary written by gemini-2.5-flash-lite from 1 source. How we write summaries →
IMPACT Introduces a new benchmark and dataset for 3D dental landmark detection, potentially accelerating research in AI-driven orthodontics.
RANK_REASON Academic paper introducing a new challenge and dataset for a specific computer vision task.