Skip to content
ICDAR 2023 BDVT-QA Competition: Born Digital Video Text Question Answering

ICDAR 2023 BDVT-QA Competition: Born Digital Video Text Question Answering

The ICDAR 2023 BDVT-QA competition, organized by Alibaba DAMO Academy, introduces a novel dataset of 1,000 born‑digital video clips for end‑to‑end video text recognition and video text question answering, offering cash prizes, detailed dataset access, and a lineup of leading academic and industry experts.

DataFunTalk
DataFunTalk
DataFunTalk

International Conference on Document Analysis and Recognition (ICDAR) is the premier conference in document analysis and recognition, gathering top researchers and industry experts every two years to discuss cutting‑edge technologies.

ICDAR 2023 BDVT‑QA (Competition for Born Digital Video Text Question Answering) is hosted by Alibaba DAMO Academy’s Text‑Vision Understanding team, focusing on the frontier of video text technology with both academic and industrial significance.

The competition features two tracks: (1) End‑to‑end video text recognition, emphasizing fusion and deduplication of video text; (2) Video text question answering, the first industry task of its kind, requiring multi‑frame text understanding.

Prize awards for each track are $2000 for first place, $1000 for second, and $500 for third.

The released dataset contains 1,000 video clips (8–60 seconds) sourced from the public internet, covering product demos, tool instructions, and film dialogues, primarily in English with some Chinese. Text annotations are polygonal, frame‑level, and include text IDs for the same object across frames; each video provides two QA pairs with reference answers.

Dataset can be downloaded from the ModelScope community (https://modelscope.cn/), searching for keywords such as "video text QA".

Organizers include senior algorithm experts and professors from Alibaba, Nanjing University, Huazhong University of Science and Technology, and the Chinese Academy of Sciences.

Sponsorship is provided by ModelScope, a leading Chinese AI model open‑source community that aggregates state‑of‑the‑art models and datasets for AI research.

Original Source

Signed-in readers can open the original source through BestHub's protected redirect.

Sign in to view source
Republication Notice

This article has been distilled and summarized from source material, then republished for learning and reference. If you believe it infringes your rights, please contactadmin@besthub.devand we will review it promptly.

AIDatasetICDARQuestion AnsweringVideo Text Recognition
Reader feedback

How this landed with the community

Sign in to like

Rate this article

Was this worth your time?

Sign in to rate
Discussion

0 Comments

Thoughtful readers leave field notes, pushback, and hard-won operational detail here.