The ICDAR 2023-DSText competition, launching on February 15, 2023, focuses on dense and small text detection and recognition in video, providing a YouTube‑sourced dataset of 100 videos, two challenge tasks, a detailed timeline, eligibility rules, and a list of international sponsoring institutions.
ShopeeVideo OCR is a multi‑language text‑recognition system for Southeast Asian e‑commerce videos that unifies detection, Transformer‑based recognition, layout analysis, and large‑scale synthetic data generation to handle Indonesian, Filipino, English, Vietnamese, Thai and Chinese scripts, delivering industry‑leading accuracy and winning thirteen ICDAR first‑place awards.
This article reviews the NeurIPS 2021 paper introducing BOVText, a large‑scale bilingual video‑text dataset with over 2,000 videos and 1.75 million frames, and describes its transformer‑based end‑to‑end video text spotter that integrates EAST encoding into DETR, covering dataset collection, annotation, architecture, and experimental results.
