magic-pdf 1.2.1__py3-none-any.whl → 1.3.0__py3-none-any.whl

magic_pdf/pre_proc/ocr_span_list_modify.py

@@ -41,6 +41,57 @@ def check_chars_is_overlap_in_span(chars):
     return False
 
 
+def remove_x_overlapping_chars(span, median_width):
+    """
+    Remove characters from a span that overlap significantly on the x-axis.
+
+    Args:
+        median_width:
+        span (dict): A span containing a list of chars, each with bbox coordinates
+                    in the format [x0, y0, x1, y1]
+
+    Returns:
+        dict: The span with overlapping characters removed
+    """
+    if 'chars' not in span or len(span['chars']) < 2:
+        return span
+
+    overlap_threshold = median_width * 0.3
+
+    i = 0
+    while i < len(span['chars']) - 1:
+        char1 = span['chars'][i]
+        char2 = span['chars'][i + 1]
+
+        # Calculate overlap width
+        x_left = max(char1['bbox'][0], char2['bbox'][0])
+        x_right = min(char1['bbox'][2], char2['bbox'][2])
+
+        if x_right > x_left:  # There is overlap
+            overlap_width = x_right - x_left
+
+            if overlap_width > overlap_threshold:
+                if char1['c'] == char2['c'] or char1['c'] == ' ' or char2['c'] == ' ':
+                    # Determine which character to remove
+                    width1 = char1['bbox'][2] - char1['bbox'][0]
+                    width2 = char2['bbox'][2] - char2['bbox'][0]
+                    if width1 < width2:
+                        # Remove the narrower character
+                        span['chars'].pop(i)
+                    else:
+                        span['chars'].pop(i + 1)
+                else:
+                    i += 1
+
+                # Don't increment i since we need to check the new pair
+            else:
+                i += 1
+        else:
+            i += 1
+
+    return span
+
+
 def remove_overlaps_min_spans(spans):
     dropped_spans = []
     #  删除重叠spans中较小的那些

magic_pdf/resources/model_config/model_configs.yaml

@@ -2,7 +2,7 @@ weights:
   layoutlmv3: Layout/LayoutLMv3/model_final.pth
   doclayout_yolo: Layout/YOLO/doclayout_yolo_docstructbench_imgsz1280_2501.pt
   yolo_v8_mfd: MFD/YOLO/yolo_v8_ft.pt
-  unimernet_small: MFR/unimernet_small_2501
+  unimernet_small: MFR/unimernet_hf_small_2503
   struct_eqtable: TabRec/StructEqTable
   tablemaster: TabRec/TableMaster
   rapid_table: TabRec/RapidTable

magic_pdf/tools/cli.py

@@ -1,15 +1,18 @@
 import os
 import shutil
 import tempfile
+from pathlib import Path
+
 import click
 import fitz
 from loguru import logger
-from pathlib import Path
 
 import magic_pdf.model as model_config
+from magic_pdf.data.batch_build_dataset import batch_build_dataset
 from magic_pdf.data.data_reader_writer import FileBasedDataReader
+from magic_pdf.data.dataset import Dataset
 from magic_pdf.libs.version import __version__
-from magic_pdf.tools.common import do_parse, parse_pdf_methods
+from magic_pdf.tools.common import batch_do_parse, do_parse, parse_pdf_methods
 from magic_pdf.utils.office_to_pdf import convert_file_to_pdf
 
 pdf_suffixes = ['.pdf']
@@ -87,37 +90,38 @@ without method specified, auto will be used by default.""",
     default=None,
 )
 def cli(path, output_dir, method, lang, debug_able, start_page_id, end_page_id):
-    model_config.__use_inside_model__ = True
-    model_config.__model_mode__ = 'full'
     os.makedirs(output_dir, exist_ok=True)
     temp_dir = tempfile.mkdtemp()
     def read_fn(path: Path):
         if path.suffix in ms_office_suffixes:
             convert_file_to_pdf(str(path), temp_dir)
-            fn = os.path.join(temp_dir, f"{path.stem}.pdf")
+            fn = os.path.join(temp_dir, f'{path.stem}.pdf')
         elif path.suffix in image_suffixes:
             with open(str(path), 'rb') as f:
                 bits = f.read()
             pdf_bytes = fitz.open(stream=bits).convert_to_pdf()
-            fn = os.path.join(temp_dir, f"{path.stem}.pdf")
+            fn = os.path.join(temp_dir, f'{path.stem}.pdf')
             with open(fn, 'wb') as f:
                 f.write(pdf_bytes)
         elif path.suffix in pdf_suffixes:
             fn = str(path)
         else:
-            raise Exception(f"Unknown file suffix: {path.suffix}")
-        
+            raise Exception(f'Unknown file suffix: {path.suffix}')
+
         disk_rw = FileBasedDataReader(os.path.dirname(fn))
         return disk_rw.read(os.path.basename(fn))
 
-    def parse_doc(doc_path: Path):
+    def parse_doc(doc_path: Path, dataset: Dataset | None = None):
         try:
             file_name = str(Path(doc_path).stem)
-            pdf_data = read_fn(doc_path)
+            if dataset is None:
+                pdf_data_or_dataset = read_fn(doc_path)
+            else:
+                pdf_data_or_dataset = dataset
             do_parse(
                 output_dir,
                 file_name,
-                pdf_data,
+                pdf_data_or_dataset,
                 [],
                 method,
                 debug_able,
@@ -130,9 +134,23 @@ def cli(path, output_dir, method, lang, debug_able, start_page_id, end_page_id):
             logger.exception(e)
 
     if os.path.isdir(path):
+        doc_paths = []
         for doc_path in Path(path).glob('*'):
             if doc_path.suffix in pdf_suffixes + image_suffixes + ms_office_suffixes:
-                parse_doc(doc_path)
+                if doc_path.suffix in ms_office_suffixes:
+                    convert_file_to_pdf(str(doc_path), temp_dir)
+                    doc_path = Path(os.path.join(temp_dir, f'{doc_path.stem}.pdf'))
+                elif doc_path.suffix in image_suffixes:
+                    with open(str(doc_path), 'rb') as f:
+                        bits = f.read()
+                        pdf_bytes = fitz.open(stream=bits).convert_to_pdf()
+                    fn = os.path.join(temp_dir, f'{doc_path.stem}.pdf')
+                    with open(fn, 'wb') as f:
+                        f.write(pdf_bytes)
+                    doc_path = Path(fn)
+                doc_paths.append(doc_path)
+        datasets = batch_build_dataset(doc_paths, 4, lang)
+        batch_do_parse(output_dir, [str(doc_path.stem) for doc_path in doc_paths], datasets, method, debug_able, lang=lang)
     else:
         parse_doc(Path(path))
 

magic_pdf/tools/common.py

@@ -8,10 +8,10 @@ import magic_pdf.model as model_config
 from magic_pdf.config.enums import SupportedPdfParseMethod
 from magic_pdf.config.make_content_config import DropMode, MakeMode
 from magic_pdf.data.data_reader_writer import FileBasedDataWriter
-from magic_pdf.data.dataset import PymuDocDataset
+from magic_pdf.data.dataset import Dataset, PymuDocDataset
 from magic_pdf.libs.draw_bbox import draw_char_bbox
-from magic_pdf.model.doc_analyze_by_custom_model import doc_analyze
-from magic_pdf.operators.models import InferenceResult
+from magic_pdf.model.doc_analyze_by_custom_model import (batch_doc_analyze,
+                                                         doc_analyze)
 
 # from io import BytesIO
 # from pypdf import PdfReader, PdfWriter
@@ -67,13 +67,13 @@ def convert_pdf_bytes_to_bytes_by_pymupdf(pdf_bytes, start_page_id=0, end_page_i
     return output_bytes
 
 
-def do_parse(
+def _do_parse(
     output_dir,
     pdf_file_name,
-    pdf_bytes,
+    pdf_bytes_or_dataset,
     model_list,
     parse_method,
-    debug_able,
+    debug_able=False,
     f_draw_span_bbox=True,
     f_draw_layout_bbox=True,
     f_dump_md=True,
@@ -92,16 +92,21 @@ def do_parse(
     formula_enable=None,
     table_enable=None,
 ):
+    from magic_pdf.operators.models import InferenceResult
     if debug_able:
         logger.warning('debug mode is on')
         f_draw_model_bbox = True
         f_draw_line_sort_bbox = True
         # f_draw_char_bbox = True
 
-    pdf_bytes = convert_pdf_bytes_to_bytes_by_pymupdf(
-        pdf_bytes, start_page_id, end_page_id
-    )
-
+    if isinstance(pdf_bytes_or_dataset, bytes):
+        pdf_bytes = convert_pdf_bytes_to_bytes_by_pymupdf(
+            pdf_bytes_or_dataset, start_page_id, end_page_id
+        )
+        ds = PymuDocDataset(pdf_bytes, lang=lang)
+    else:
+        ds = pdf_bytes_or_dataset
+    pdf_bytes = ds._raw_data
     local_image_dir, local_md_dir = prepare_env(output_dir, pdf_file_name, parse_method)
 
     image_writer, md_writer = FileBasedDataWriter(local_image_dir), FileBasedDataWriter(
@@ -109,8 +114,6 @@ def do_parse(
     )
     image_dir = str(os.path.basename(local_image_dir))
 
-    ds = PymuDocDataset(pdf_bytes, lang=lang)
-
     if len(model_list) == 0:
         if model_config.__use_inside_model__:
             if parse_method == 'auto':
@@ -241,5 +244,80 @@ def do_parse(
 
     logger.info(f'local output dir is {local_md_dir}')
 
+def do_parse(
+    output_dir,
+    pdf_file_name,
+    pdf_bytes_or_dataset,
+    model_list,
+    parse_method,
+    debug_able=False,
+    f_draw_span_bbox=True,
+    f_draw_layout_bbox=True,
+    f_dump_md=True,
+    f_dump_middle_json=True,
+    f_dump_model_json=True,
+    f_dump_orig_pdf=True,
+    f_dump_content_list=True,
+    f_make_md_mode=MakeMode.MM_MD,
+    f_draw_model_bbox=False,
+    f_draw_line_sort_bbox=False,
+    f_draw_char_bbox=False,
+    start_page_id=0,
+    end_page_id=None,
+    lang=None,
+    layout_model=None,
+    formula_enable=None,
+    table_enable=None,
+):
+    parallel_count = 1
+    if os.environ.get('MINERU_PARALLEL_INFERENCE_COUNT'):
+        parallel_count = int(os.environ['MINERU_PARALLEL_INFERENCE_COUNT'])
+
+    if parallel_count > 1:
+        if isinstance(pdf_bytes_or_dataset, bytes):
+            pdf_bytes = convert_pdf_bytes_to_bytes_by_pymupdf(
+                pdf_bytes_or_dataset, start_page_id, end_page_id
+            )
+            ds = PymuDocDataset(pdf_bytes, lang=lang)
+        else:
+            ds = pdf_bytes_or_dataset
+        batch_do_parse(output_dir, [pdf_file_name], [ds], parse_method, debug_able, f_draw_span_bbox=f_draw_span_bbox, f_draw_layout_bbox=f_draw_layout_bbox, f_dump_md=f_dump_md, f_dump_middle_json=f_dump_middle_json, f_dump_model_json=f_dump_model_json, f_dump_orig_pdf=f_dump_orig_pdf, f_dump_content_list=f_dump_content_list, f_make_md_mode=f_make_md_mode, f_draw_model_bbox=f_draw_model_bbox, f_draw_line_sort_bbox=f_draw_line_sort_bbox, f_draw_char_bbox=f_draw_char_bbox, lang=lang)
+    else:
+        _do_parse(output_dir, pdf_file_name, pdf_bytes_or_dataset, model_list, parse_method, debug_able, start_page_id=start_page_id, end_page_id=end_page_id, lang=lang, layout_model=layout_model, formula_enable=formula_enable, table_enable=table_enable,  f_draw_span_bbox=f_draw_span_bbox, f_draw_layout_bbox=f_draw_layout_bbox, f_dump_md=f_dump_md, f_dump_middle_json=f_dump_middle_json, f_dump_model_json=f_dump_model_json, f_dump_orig_pdf=f_dump_orig_pdf, f_dump_content_list=f_dump_content_list, f_make_md_mode=f_make_md_mode, f_draw_model_bbox=f_draw_model_bbox, f_draw_line_sort_bbox=f_draw_line_sort_bbox, f_draw_char_bbox=f_draw_char_bbox)
+
+
+def batch_do_parse(
+    output_dir,
+    pdf_file_names: list[str],
+    pdf_bytes_or_datasets: list[bytes | Dataset],
+    parse_method,
+    debug_able=False,
+    f_draw_span_bbox=True,
+    f_draw_layout_bbox=True,
+    f_dump_md=True,
+    f_dump_middle_json=True,
+    f_dump_model_json=True,
+    f_dump_orig_pdf=True,
+    f_dump_content_list=True,
+    f_make_md_mode=MakeMode.MM_MD,
+    f_draw_model_bbox=False,
+    f_draw_line_sort_bbox=False,
+    f_draw_char_bbox=False,
+    lang=None,
+    layout_model=None,
+    formula_enable=None,
+    table_enable=None,
+):
+    dss = []
+    for v in pdf_bytes_or_datasets:
+        if isinstance(v, bytes):
+            dss.append(PymuDocDataset(v, lang=lang))
+        else:
+            dss.append(v)
+
+    infer_results = batch_doc_analyze(dss, parse_method, lang=lang, layout_model=layout_model, formula_enable=formula_enable, table_enable=table_enable)
+    for idx, infer_result in enumerate(infer_results):
+        _do_parse(output_dir, pdf_file_names[idx], dss[idx], infer_result.get_infer_res(), parse_method, debug_able, f_draw_span_bbox=f_draw_span_bbox, f_draw_layout_bbox=f_draw_layout_bbox, f_dump_md=f_dump_md, f_dump_middle_json=f_dump_middle_json, f_dump_model_json=f_dump_model_json, f_dump_orig_pdf=f_dump_orig_pdf, f_dump_content_list=f_dump_content_list, f_make_md_mode=f_make_md_mode, f_draw_model_bbox=f_draw_model_bbox, f_draw_line_sort_bbox=f_draw_line_sort_bbox, f_draw_char_bbox=f_draw_char_bbox, lang=lang)
+
 
 parse_pdf_methods = click.Choice(['ocr', 'txt', 'auto'])

{magic_pdf-1.2.1.dist-info → magic_pdf-1.3.0.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: magic-pdf
-Version: 1.2.1
+Version: 1.3.0
 Summary: A practical tool for converting PDF to Markdown
 Home-page: https://github.com/opendatalab/MinerU
 Requires-Python: >=3.9
@@ -9,35 +9,30 @@ License-File: LICENSE.md
 Requires-Dist: boto3>=1.28.43
 Requires-Dist: Brotli>=1.1.0
 Requires-Dist: click>=8.1.7
-Requires-Dist: fast-langdetect>=0.2.3
+Requires-Dist: fast-langdetect<0.3.0,>=0.2.3
 Requires-Dist: loguru>=0.6.0
-Requires-Dist: numpy<2.0.0,>=1.21.6
-Requires-Dist: pydantic>=2.7.2
-Requires-Dist: PyMuPDF<=1.24.14,>=1.24.9
+Requires-Dist: numpy>=1.21.6
+Requires-Dist: pydantic<2.11,>=2.7.2
+Requires-Dist: PyMuPDF<1.25.0,>=1.24.9
 Requires-Dist: scikit-learn>=1.0.2
-Requires-Dist: torch>=2.2.2
-Requires-Dist: transformers
+Requires-Dist: torch!=2.5.0,!=2.5.1,<=2.6.0,>=2.2.2
+Requires-Dist: torchvision
+Requires-Dist: transformers<5.0.0,>=4.49.0
 Requires-Dist: pdfminer.six==20231228
+Requires-Dist: tqdm>=4.67.1
 Provides-Extra: full
-Requires-Dist: unimernet==0.2.3; extra == "full"
-Requires-Dist: torch<=2.3.1,>=2.2.2; extra == "full"
-Requires-Dist: torchvision<=0.18.1,>=0.17.2; extra == "full"
 Requires-Dist: ultralytics>=8.3.48; extra == "full"
-Requires-Dist: paddleocr==2.7.3; extra == "full"
-Requires-Dist: struct-eqtable==0.3.2; extra == "full"
-Requires-Dist: einops; extra == "full"
-Requires-Dist: accelerate; extra == "full"
 Requires-Dist: doclayout-yolo==0.0.2b1; extra == "full"
-Requires-Dist: rapidocr-paddle<2.0.0,>=1.4.5; extra == "full"
-Requires-Dist: rapidocr-onnxruntime<2.0.0,>=1.4.4; extra == "full"
+Requires-Dist: dill<1,>=0.3.9; extra == "full"
 Requires-Dist: rapid-table<2.0.0,>=1.0.3; extra == "full"
-Requires-Dist: PyYAML; extra == "full"
-Requires-Dist: openai; extra == "full"
-Requires-Dist: detectron2; extra == "full"
-Requires-Dist: matplotlib; (platform_system == "Linux" or platform_system == "Darwin") and extra == "full"
-Requires-Dist: paddlepaddle==3.0.0rc1; (platform_system == "Linux" or platform_system == "Darwin") and extra == "full"
+Requires-Dist: PyYAML<7,>=6.0.2; extra == "full"
+Requires-Dist: ftfy<7,>=6.3.1; extra == "full"
+Requires-Dist: openai<2,>=1.70.0; extra == "full"
+Requires-Dist: shapely<3,>=2.0.7; extra == "full"
+Requires-Dist: pyclipper<2,>=1.3.0; extra == "full"
+Requires-Dist: omegaconf<3,>=2.3.0; extra == "full"
+Requires-Dist: matplotlib>=3.10; (platform_system == "Linux" or platform_system == "Darwin") and extra == "full"
 Requires-Dist: matplotlib<=3.9.0; platform_system == "Windows" and extra == "full"
-Requires-Dist: paddlepaddle==2.6.1; platform_system == "Windows" and extra == "full"
 Provides-Extra: lite
 Requires-Dist: paddleocr==2.7.3; extra == "lite"
 Requires-Dist: paddlepaddle==3.0.0b1; platform_system == "Linux" and extra == "lite"
@@ -94,6 +89,23 @@ Easier to use: Just grab MinerU Desktop. No coding, no login, just a simple inte
 </div>
 
 # Changelog
+- 2025/04/03 Release of 1.3.0, in this version we made many optimizations and improvements:
+  - Installation and compatibility optimization
+    - By removing the use of `layoutlmv3` in layout, resolved compatibility issues caused by `detectron2`.
+    - Torch version compatibility extended to 2.2~2.6 (excluding 2.5).
+    - CUDA compatibility supports 11.8/12.4/12.6 (CUDA version determined by torch), resolving compatibility issues for some users with 50-series and H-series GPUs.
+    - Python compatible versions expanded to 3.10~3.12, solving the problem of automatic downgrade to 0.6.1 during installation in non-3.10 environments.
+    - Offline deployment process optimized; no internet connection required after successful deployment to download any model files.
+  - Performance optimization
+    - By supporting batch processing of multiple PDF files ([script example](demo/batch_demo.py)), improved parsing speed for small files in batches (compared to version 1.0.1, formula parsing speed increased by over 1400%, overall parsing speed increased by over 500%).
+    - Optimized loading and usage of the mfr model, reducing GPU memory usage and improving parsing speed (requires re-execution of the [model download process](docs/how_to_download_models_en.md) to obtain incremental updates of model files).
+    - Optimized GPU memory usage, requiring only a minimum of 6GB to run this project.
+    - Improved running speed on MPS devices.
+  - Parsing effect optimization
+    - Updated the mfr model to `unimernet(2503)`, solving the issue of lost line breaks in multi-line formulas.
+  - Usability Optimization
+    - By using `paddleocr2torch`, completely replaced the use of the `paddle` framework and `paddleocr` in the project, resolving conflicts between `paddle` and `torch`, as well as thread safety issues caused by the `paddle` framework.
+    - Added a real-time progress bar during the parsing process to accurately track progress, making the wait less painful.
 - 2025/03/03 1.2.1 released, fixed several bugs:
   - Fixed the impact on punctuation marks during full-width to half-width conversion of letters and numbers
   - Fixed caption matching inaccuracies in certain scenarios
@@ -262,7 +274,7 @@ There are three different ways to experience MinerU:
     </tr>
     <tr>
         <td colspan="3">Python Version</td>
-        <td colspan="3">3.10(Please make sure to create a Python 3.10 virtual environment using conda)</td>
+        <td colspan="3">3.10~3.12</td>
     </tr>
     <tr>
         <td colspan="3">Nvidia Driver Version</td>
@@ -272,8 +284,8 @@ There are three different ways to experience MinerU:
     </tr>
     <tr>
         <td colspan="3">CUDA Environment</td>
-        <td>Automatic installation [12.1 (pytorch) + 11.8 (paddle)]</td>
-        <td>11.8 (manual installation) + cuDNN v8.7.0 (manual installation)</td>
+        <td>11.8/12.4/12.6</td>
+        <td>11.8/12.4/12.6</td>
         <td>None</td>
     </tr>
     <tr>
@@ -283,11 +295,11 @@ There are three different ways to experience MinerU:
         <td>None</td>
     </tr>
     <tr>
-        <td rowspan="2">GPU Hardware Support List</td>
-        <td colspan="2">GPU VRAM 8GB or more</td>
-        <td colspan="2">2080~2080Ti / 3060Ti~3090Ti / 4060~4090<br>
-        8G VRAM can enable all acceleration features</td>
-        <td rowspan="2">None</td>
+        <td rowspan="2">GPU/MPS Hardware Support List</td>
+        <td colspan="2">GPU VRAM 6GB or more</td>
+        <td colspan="2">All GPUs with Tensor Cores produced from Volta(2017) onwards.<br>
+        More than 6GB VRAM </td>
+        <td rowspan="2">apple slicon</td>
     </tr>
 </table>
 
@@ -304,9 +316,9 @@ Synced with dev branch updates:
 #### 1. Install magic-pdf
 
 ```bash
-conda create -n mineru python=3.10
+conda create -n mineru 'python<3.13' -y
 conda activate mineru
-pip install -U "magic-pdf[full]" --extra-index-url https://wheels.myhloli.com
+pip install -U "magic-pdf[full]"
 ```
 
 #### 2. Download model weight files
@@ -331,7 +343,7 @@ You can modify certain configurations in this file to enable or disable features
 {
     // other config
     "layout-config": {
-        "model": "doclayout_yolo" // Please change to "layoutlmv3" when using layoutlmv3.
+        "model": "doclayout_yolo" 
     },
     "formula-config": {
         "mfd_model": "yolo_v8_mfd",
@@ -339,8 +351,8 @@ You can modify certain configurations in this file to enable or disable features
         "enable": true  // The formula recognition feature is enabled by default. If you need to disable it, please change the value here to "false".
     },
     "table-config": {
-        "model": "rapid_table",  // Default to using "rapid_table", can be switched to "tablemaster" or "struct_eqtable".
-        "sub_model": "slanet_plus",  // When the model is "rapid_table", you can choose a sub_model. The options are "slanet_plus" and "unitable"
+        "model": "rapid_table", 
+        "sub_model": "slanet_plus",
         "enable": true, // The table recognition feature is enabled by default. If you need to disable it, please change the value here to "false".
         "max_time": 400
     }
@@ -355,7 +367,7 @@ If your device supports CUDA and meets the GPU requirements of the mainline envi
 - [Windows 10/11 + GPU](docs/README_Windows_CUDA_Acceleration_en_US.md)
 - Quick Deployment with Docker
 > [!IMPORTANT]
-> Docker requires a GPU with at least 8GB of VRAM, and all acceleration features are enabled by default.
+> Docker requires a GPU with at least 6GB of VRAM, and all acceleration features are enabled by default.
 >
 > Before running this Docker, you can use the following command to check if your device supports CUDA acceleration on Docker.
 > 
@@ -377,7 +389,7 @@ If your device has NPU acceleration hardware, you can follow the tutorial below
 
 ### Using MPS
 
-If your device uses Apple silicon chips, you can enable MPS acceleration for certain supported tasks (such as layout detection and formula detection).
+If your device uses Apple silicon chips, you can enable MPS acceleration for your tasks.
 
 You can enable MPS acceleration by setting the `device-mode` parameter to `mps` in the `magic-pdf.json` configuration file.
 
@@ -388,10 +400,6 @@ You can enable MPS acceleration by setting the `device-mode` parameter to `mps`
 }
 ```
 
-> [!TIP]
-> Since the formula recognition task cannot utilize MPS acceleration, you can disable the formula recognition feature in tasks where it is not needed to achieve optimal performance.
->
-> You can disable the formula recognition feature by setting the `enable` parameter in the `formula-config` section to `false`.
 
 ## Usage
 
@@ -465,6 +473,8 @@ This project currently uses PyMuPDF to achieve advanced functionality. However,
 - [StructEqTable](https://github.com/UniModal4Reasoning/StructEqTable-Deploy)
 - [RapidTable](https://github.com/RapidAI/RapidTable)
 - [PaddleOCR](https://github.com/PaddlePaddle/PaddleOCR)
+- [RapidOCR](https://github.com/RapidAI/RapidOCR)
+- [PaddleOCR2Pytorch](https://github.com/frotms/PaddleOCR2Pytorch)
 - [PyMuPDF](https://github.com/pymupdf/PyMuPDF)
 - [layoutreader](https://github.com/ppaanngggg/layoutreader)
 - [fast-langdetect](https://github.com/LlmKira/fast-langdetect)