ONNX Model Zoo

AXERA-TECH/PPOCR_v5

By AXERA-TECH · ↓ 4 · ♥ 2 · Updated 2025-11-13

What's in this repo

This page lists the files in AXERA-TECH/PPOCR_v5, including cls_mobile_sim_static.onnx, det_mobile_sim_static.onnx, rec_mobile_sim_static.onnx.

  • 3 ONNX models (.onnx files)
  • 29 files, including 12 .axmodel, 4 .json, 3 .jpg, 3 .onnx, 2 .py, 2 .txt
  • Last updated 2025-11-13

ONNX files

File Size Download
cls_mobile_sim_static.onnx — Download
det_mobile_sim_static.onnx — Download
rec_mobile_sim_static.onnx — Download

How to use

These ONNX files can be loaded with ONNX Runtime. To inspect a model graph, upload the file to the Netron viewer. Download links point to the original files on Hugging Face.

README

PPOCR_v5

This version of PPOCR_v5 has been converted to run on the Axera NPU using w8a16 quantization.

This model has been optimized with the following LoRA:

Compatible with Pulsar2 version: 4.1-patch1

Convert tools links:

For those who are interested in model conversion, you can try to export axmodel through

Support Platform

Chips model cost
PP-OCRv5_mobile_det 16.8ms
AX650 PP-OCRv5_mobile_rec 1.4ms
PP-LCNet_x0_25_textline_ori 0.17ms
PP-OCRv5_mobile_det 121.7ms
AX630C PP-OCRv5_mobile_rec 17.8ms
PP-LCNet_x0_25_textline_ori 0.9ms

How to use

Download all files from this repository to the device


root@ax650:~/PPOCR_v5# tree
.
|-- ax650
|   `-- det.axmodel
|   `-- rec.axmodel
|   `-- cls.axmodel
|-- infer_onnx.py
|-- infer_axmodel.py
|-- 11.jpg
`-- res_onnx.jpg
`-- res_ax.jpg

Inference

Input image:

Inference with AX650 Host, such as M4N-Dock(爱芯派Pro)
root@ax650:~/PPOCR_v5# python3 infer_axmodel.py
[INFO] Available providers:  ['AxEngineExecutionProvider']
[INFO] Using provider: AxEngineExecutionProvider
[INFO] Chip type: ChipType.MC50
[INFO] VNPU type: VNPUType.DISABLED
[INFO] Engine version: 2.12.0s
[INFO] Model type: 2 (triple core)
[INFO] Compiler version: 4.2 b98901c3
[INFO] Using provider: AxEngineExecutionProvider
[INFO] Model type: 0 (single core)
[INFO] Compiler version: 4.2 b98901c3
[INFO] Using provider: AxEngineExecutionProvider
[INFO] Model type: 2 (triple core)
[INFO] Compiler version: 4.2 b98901c3
[[[22.0, 32.0], [307.0, 32.0], [307.0, 77.0], [22.0, 77.0]], ('纯臻营养护发素', 0.9881175756454468)]
[[[24.0, 80.0], [174.0, 80.0], [174.0, 105.0], [24.0, 105.0]], ('产品信息/参数', 0.9941993951797485)]
[[[25.0, 110.0], [333.0, 110.0], [333.0, 135.0], [25.0, 135.0]], ('(45元/每公斤,100公斤起订)', 0.9549185633659363)]
[[[24.0, 140.0], [283.0, 142.0], [283.0, 167.0], [24.0, 165.0]], ('每瓶22元,1000瓶起订)', 0.9522098898887634)]
[[[23.0, 175.0], [302.0, 174.0], [302.0, 197.0], [23.0, 198.0]], ('【品牌】:代加工方式/OEMODM', 0.98729008436203)]
[[[24.0, 206.0], [235.0, 208.0], [235.0, 229.0], [24.0, 228.0]], ('【品名】:纯臻营养护发素', 0.9895920753479004)]
[[[26.0, 238.0], [242.0, 238.0], [242.0, 259.0], [26.0, 259.0]], ('【产品编号】:YM-X-3011', 0.9853904843330383)]
[[[412.0, 233.0], [430.0, 233.0], [430.0, 303.0], [412.0, 303.0]], ('ODMOEM', 0.8154547810554504)]
[[[24.0, 269.0], [181.0, 268.0], [181.0, 289.0], [25.0, 290.0]], ('【净含量】:220ml', 0.9779437780380249)]
[[[24.0, 300.0], [253.0, 301.0], [253.0, 322.0], [24.0, 321.0]], ('【适用人群】:适合所有肤质', 0.9893492460250854)]
[[[24.0, 331.0], [345.0, 332.0], [345.0, 354.0], [24.0, 352.0]], ('【主要成分】:鲸蜡硬脂醇、燕麦β-葡聚', 0.9715114235877991)]
[[[26.0, 364.0], [283.0, 364.0], [283.0, 384.0], [26.0, 384.0]], ('糖、椰油酰胺丙基甜菜碱、泛醌', 0.9863923192024231)]
[[[367.0, 366.0], [476.0, 366.0], [476.0, 388.0], [367.0, 388.0]], ('(成品包材)', 0.8738409876823425)]
[[[25.0, 394.0], [362.0, 394.0], [362.0, 416.0], [25.0, 416.0]], ('【主要功能】:可紧致头发磷层,从而达到', 0.9911136627197266)]
[[[27.0, 426.0], [373.0, 426.0], [373.0, 447.0], [27.0, 447.0]], ('即时持久改善头发光泽的效果,给干燥的头', 0.9935663342475891)]
[[[26.0, 457.0], [137.0, 457.0], [137.0, 478.0], [26.0, 478.0]], ('发足够的滋养', 0.9879047274589539)]

Output image:

Model info

onnx OCR paddle PPOCRv5 text-classification en license:mit region:us
↓ 4 ♥ 2

Created: 2025-07-24

Updated: 2025-11-13

View on Hugging Face

Files (29)

.gitattributes
11.jpg
README.md
ax620e/cls_npu1.axmodel
ax620e/cls_npu2.axmodel
ax620e/det_npu1.axmodel
ax620e/det_npu2.axmodel
ax620e/rec_npu1.axmodel
ax620e/rec_npu2.axmodel
ax650/cls_npu1.axmodel
ax650/cls_npu3.axmodel
ax650/det_npu1.axmodel
ax650/det_npu3.axmodel
ax650/rec_npu1.axmodel
ax650/rec_npu3.axmodel
cls.json
cls_mobile_sim_static.onnx ONNX
config.json
det.json
det_mobile_sim_static.onnx ONNX
infer_axmodel.py
infer_onnx.py
ppocrv5_dict.txt
rec.json
rec_mobile_sim_static.onnx ONNX
requirements.txt
res_ax.jpg
res_onnx.jpg
simfang.ttf

Related models