Safetensors
rt_detr_v2
nlivathinos commited on
Commit
1907ed0
·
verified ·
1 Parent(s): bdb7099

Add technical report link and text improvements (#8)

Browse files

- docs: Add technical report link and text improvements (6b580afc0fc91926eab346428da1a943a792c398)

Files changed (2) hide show
  1. README.md +21 -18
  2. docling_heron_400.png +0 -0
README.md CHANGED
@@ -2,17 +2,21 @@
2
  license: apache-2.0
3
  ---
4
 
5
- THIS IS WORK IN PROGRESS
6
 
 
7
 
8
- # Docling Layout Model
9
 
10
- `docling-layout-heron` is the Layout Model of [Docling project](https://github.com/docling-project/docling).
11
 
12
- This model uses the [RT-DETRv2](https://github.com/lyuwenyu/RT-DETR/tree/main/rtdetrv2_pytorch) architecture and has been trained from scratch on a variety of document datasets.
 
 
13
 
14
 
15
- # Inference code example
 
16
 
17
  Prerequisites:
18
 
@@ -83,9 +87,19 @@ for result in results:
83
  ```
84
 
85
 
86
- # References
87
 
88
  ```
 
 
 
 
 
 
 
 
 
 
89
  @techreport{Docling,
90
  author = {Deep Search Team},
91
  month = {8},
@@ -96,15 +110,4 @@ for result in results:
96
  version = {1.0.0},
97
  year = {2024}
98
  }
99
-
100
- @misc{lv2024rtdetrv2improvedbaselinebagoffreebies,
101
- title={RT-DETRv2: Improved Baseline with Bag-of-Freebies for Real-Time Detection Transformer},
102
- author={Wenyu Lv and Yian Zhao and Qinyao Chang and Kui Huang and Guanzhong Wang and Yi Liu},
103
- year={2024},
104
- eprint={2407.17140},
105
- archivePrefix={arXiv},
106
- primaryClass={cs.CV},
107
- url={https://arxiv.org/abs/2407.17140},
108
- }
109
-
110
- ```
 
2
  license: apache-2.0
3
  ---
4
 
5
+ ![heron_logo](docling_heron_400.png)
6
 
7
+ # Document Layout Analysis "heron"
8
 
9
+ 🚀 **`heron`** is the **default layout analysis model** of the [Docling project](https://github.com/docling-project/docling), designed for robust and high-quality document layout understanding.
10
 
11
+ 📄 For an in-depth description of the **model architecture**, **training datasets**, and **evaluation methodology**, please refer to our technical report:
12
 
13
+ **Advanced Layout Analysis Models for Docling**
14
+ Nikolaos Livathinos *et al.*
15
+ [🔗 https://arxiv.org/abs/2509.11720](https://arxiv.org/abs/2509.11720)
16
 
17
 
18
+
19
+ ## Inference code example
20
 
21
  Prerequisites:
22
 
 
87
  ```
88
 
89
 
90
+ ## References
91
 
92
  ```
93
+ @misc{livathinos2025advancedlayoutanalysismodels,
94
+ title={advanced layout analysis models for docling},
95
+ author={nikolaos livathinos and christoph auer and ahmed nassar and rafael teixeira de lima and maksym lysak and brown ebouky and cesar berrospi and michele dolfi and panagiotis vagenas and matteo omenetti and kasper dinkla and yusik kim and valery weber and lucas morin and ingmar meijer and viktor kuropiatnyk and tim strohmeyer and a. said gurbuz and peter w. j. staar},
96
+ year={2025},
97
+ eprint={2509.11720},
98
+ archiveprefix={arxiv},
99
+ primaryclass={cs.cv},
100
+ url={https://arxiv.org/abs/2509.11720},
101
+ }
102
+
103
  @techreport{Docling,
104
  author = {Deep Search Team},
105
  month = {8},
 
110
  version = {1.0.0},
111
  year = {2024}
112
  }
113
+ ```
 
 
 
 
 
 
 
 
 
 
 
docling_heron_400.png ADDED