Paddle-StructureV3产线的识别结果中，PDF文档中的图片发生扭曲

[refrainD]

🔎 Search before asking

• I have searched the PaddleOCR Docs and found no similar bug report.
• I have searched the PaddleOCR Issues and found no similar bug report.
• I have searched the PaddleOCR Discussions and found no similar bug report.

🐛 Bug (问题描述)

解析结果中的图片发生扭曲。

🏃‍♂️ Environment (运行环境)

docker镜像: ccr-2vdh3abv-pub.cnc.bj.baidubce.com/paddlepaddle/paddle:3.2.1-gpu-cuda13.0-cudnn9.13

🌰 Minimal Reproducible Example (最小可复现问题的Demo)

基础服务化部署: paddlex --serve --pipeline PP-StructureV3
使用文档的Python集成脚本调用，解析PDF文件，PDF中的图片在解析后发生扭曲。

[elsonwx]

same issue

paddlepaddle-gpu==3.3.0
paddleocr[all] latest version
cuda version 12.9