Tavondo
CompetitionsEventsGuidesPlatforms
中文Start matching
CompetitionsEventsGuidesPlatforms
中文Sign in
Tavondo

Find AI competitions and events worth your time.

CompetitionsEventsGuidesPlatforms
Sign inRegisterStart matching

© 2026 Tavondo

中文
  1. Home
  2. Competitions
  3. PaddleOCR 算法模型挑战赛 - 赛题二:通用表格识别任务
Baidu AI StudioComputer VisionIntermediateRegistration unverified

PaddleOCR 算法模型挑战赛 - 赛题二:通用表格识别任务

A intermediate computer vision competition requiring roughly 10–30 hours for an end-to-end practice run.

Difficulty
Intermediate
Estimated effort
10–30 hours
Registration deadline
To be confirmed
Compute
Single GPU
OverviewPreparationSource

Overview

Who this competition fits

适合已有 PaddlePaddle/PaddleOCR 使用经验、能训练和优化图像表格结构识别模型的团队。参赛者还需要能处理 PubTabNet 表格图像,并按指定 JSON 格式生成 HTML 结构预测;若希望进入决赛,还应能在 AtomGit 开源完整训练、推理代码和权重,并参加线下答辩。

Read the original official blurb
PaddleOCR 旨在打造一套丰富、领先、且实用的 OCR 工具库,助力开发者训练出更好的模型,并应用落地。作为飞桨开源社区最热门的套件开源项目,PaddleOCR 拥有业界领先的 PP-OCR 模型和多场景通用的表格识别模型,广受众多开发者和用户的认可和使用。本次算法模型挑战赛聚焦通用 OCR 模型,要求开发者以当前飞桨开源模型为基线,不断突破技术瓶颈,研发出更加高效准确的新模型。 Official tags: 通用 OCR 模型. Sign-ups: 245.

Preparation

From registration to a first submission

  1. 01

    PaddlePaddle

  2. 02

    PaddleOCR

  3. 03

    表格结构识别

  4. 04

    PubTabNet 数据集

  5. 05

    HTML 表格序列生成

  6. 06

    JSONL 结果文件生成

Before you commit: 核心难点是同时提高复杂表格结构的逐 Token 识别正确性并维持接近基线的推理速度:单个 Token 出错即会使该图像判错。提交不仅是跑出预测结果,还要求使用 PaddlePaddle 训练,并在复核时提供可复现最佳成绩的源码;进入决赛的队伍还需开源代码。

Source

How this page was assembled

Competition information is structured from the official page. Scores are platform estimates for decision support; official rules take precedence.

Official competition page
Baidu AI Studio
Last checked
Not recorded

Entry assessment

est.
55

Prepare before joining

Overall fit estimate

Beginner fit55/100
Learning value58/100

Confirm eligibility and time commitment before moving into formal preparation.

Entry eligibility needs official confirmation

Open official pageSign in to saveOpen guide

Data and review state

Separate what is confirmed from what still needs review. Official rules and deadlines win.

View data stateClose
Source page verifiedMetric confirmedReviewed
Published on this site
Jul 21, 2026

Official metric is not published on the Baidu AI Studio competition listing; verify scoring on the competition page.