Change Title link (#2)
Browse files- Change Title link (76081691fb97261dc4f579467a9b9da6be82320b)
Co-authored-by: Ethannwan <EthannW@users.noreply.huggingface.co>
README.md
CHANGED
|
@@ -7,12 +7,6 @@ pipeline_tag: image-text-to-text
|
|
| 7 |
library_name: transformers
|
| 8 |
---
|
| 9 |
|
| 10 |
-
<div align="center">
|
| 11 |
-
|
| 12 |
-
# HunyuanOCR
|
| 13 |
-
|
| 14 |
-
</div>
|
| 15 |
-
|
| 16 |
<p align="center">
|
| 17 |
<img src="https://github.com/Tencent-Hunyuan/HunyuanOCR/blob/main/assets/hyocr-head-img.png?raw=true" width="80%"/> <br>
|
| 18 |
</p>
|
|
@@ -25,6 +19,12 @@ library_name: transformers
|
|
| 25 |
<a href="https://github.com/Tencent-Hunyuan/HunyuanOCR"><b>🌟 Github</b></a>
|
| 26 |
</p>
|
| 27 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 28 |
|
| 29 |
## 📖 Introduction
|
| 30 |
**HunyuanOCR** stands as a leading end-to-end OCR expert VLM powered by Hunyuan's native multimodal architecture. With a remarkably lightweight 1B parameter design, it has achieved multiple state-of-the-art benchmarks across the industry. The model demonstrates mastery in **complex multilingual document parsing** while excelling in practical applications including **text spotting, open-field information extraction, video subtitle extraction, and photo translation**.
|
|
|
|
| 7 |
library_name: transformers
|
| 8 |
---
|
| 9 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 10 |
<p align="center">
|
| 11 |
<img src="https://github.com/Tencent-Hunyuan/HunyuanOCR/blob/main/assets/hyocr-head-img.png?raw=true" width="80%"/> <br>
|
| 12 |
</p>
|
|
|
|
| 19 |
<a href="https://github.com/Tencent-Hunyuan/HunyuanOCR"><b>🌟 Github</b></a>
|
| 20 |
</p>
|
| 21 |
|
| 22 |
+
<h2>
|
| 23 |
+
<p align="center">
|
| 24 |
+
<a href="https://github.com/Tencent-Hunyuan/HunyuanOCR/blob/main/HunyuanOCR_Technical_Report.pdf">HunyuanOCR</a>
|
| 25 |
+
</p>
|
| 26 |
+
</h2>
|
| 27 |
+
|
| 28 |
|
| 29 |
## 📖 Introduction
|
| 30 |
**HunyuanOCR** stands as a leading end-to-end OCR expert VLM powered by Hunyuan's native multimodal architecture. With a remarkably lightweight 1B parameter design, it has achieved multiple state-of-the-art benchmarks across the industry. The model demonstrates mastery in **complex multilingual document parsing** while excelling in practical applications including **text spotting, open-field information extraction, video subtitle extraction, and photo translation**.
|