EasyEdge PaddleOCR v4?

[gutenye]

Only PaddleOCR v3 is available in the https://ai.baidu.com/easyedge/app/openSource?from=paddlelite

1. Is v4 available for download?
2. If not, how can I replace the v3 demo with v4 model?

[UserWangZz]

https://github.com/PaddlePaddle/PaddleOCR#%EF%B8%8F-pp-ocr%E7%B3%BB%E5%88%97%E6%A8%A1%E5%9E%8B%E5%88%97%E8%A1%A8%E6%9B%B4%E6%96%B0%E4%B8%AD
you can download ppocrv4 model in this link.

[gutenye]

@UserWangZz

The PP-OCRv3 iOS SDK I downloaded from https://ai.baidu.com/easyedge/app/openSource?from=paddlelite has two files model and params

The ppocrv4 models from your link have four files (detection and recognition)

ch_PP-OCRv4_det_infer/inference.pdmodel
ch_PP-OCRv4_det_infer/inference.pdiparams
ch_PP-OCRv4_rec_infer/inference.pdmodel
ch_PP-OCRv4_rec_infer/inference.pdiparams

How can I replace the two files (PP-OCRv3 iOS SDK) with the four files (ppocrv4)?

[tink2123]

Yes, you can directly replace the model file with ppocr_v4

[gutenye]

@tink2123 the SDK only has two files, while ppocr_v4 has four files. Which files should I use?