Beginner score
78/100
Competition workspace
适合已有计算机视觉目标检测基础、愿意使用 PaddlePaddle 2.0 及以上版本训练端到端模型的个人参赛者。你需要能处理钢铁表面灰度图像，并输出缺陷类别、边界框坐标和置信度。
Suggested next step
Decide whether this contest fits your current stage before you sink time into the leaderboard.
Beginner score
78/100
Learning value
72/100
Estimated effort
4-12 hours
Metric
Official scoring is published on the Ba…
适合已有计算机视觉目标检测基础、愿意使用 PaddlePaddle 2.0 及以上版本训练端到端模型的个人参赛者。你需要能处理钢铁表面灰度图像，并输出缺陷类别、边界框坐标和置信度。
Taken from the competition's official page.
If the items below still feel unfamiliar, you usually get a better result by preparing first instead of rushing in.
PaddlePaddle 2.0+
目标检测模型
边界框标注与坐标
图像缺陷分类
CSV结果文件生成
mAP与IoU评估
The real friction is usually not library usage. It is validation, time allocation, and task framing.
难点在于同时做好缺陷定位和六类缺陷分类：同类缺陷的形态、方向和灰度会变化，而不同类别之间也需要区分。提交还必须生成严格格式的 submission.csv，且每个检测框都要单独成行。
This guide is the best pre-read if you want a cleaner start instead of trial-and-error.
A pre-flight checklist for image competitions when you only have free cloud GPU access.
Your First CV Baseline ChecklistUse these fields to make a quick decision before you dive deeper.
Registration is unverified; confirm availability on the official page before investing.
Sign in to save competitions and build your own shortlist.
Rules, files, submission details, and the live deadline still come from the official page.
These competitions share a similar domain or difficulty level.
Kaggle
Computer VisionKaggle
Computer VisionDrivenData
Computer VisionDrivenData
Computer VisionSeparate what is confirmed from what still needs review. Official rules and deadlines win — report anything that looks wrong.
Official metric is not published on the Baidu AI Studio competition listing; verify scoring on the competition page.