poster2json 0.1.0__tar.gz

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (15) hide show
  1. poster2json-0.1.0/LICENSE.md +22 -0
  2. poster2json-0.1.0/PKG-INFO +263 -0
  3. poster2json-0.1.0/README.md +218 -0
  4. poster2json-0.1.0/poster2json/__init__.py +47 -0
  5. poster2json-0.1.0/poster2json/__main__.py +8 -0
  6. poster2json-0.1.0/poster2json/cli.py +272 -0
  7. poster2json-0.1.0/poster2json/extract.py +933 -0
  8. poster2json-0.1.0/poster2json/gui.py +37 -0
  9. poster2json-0.1.0/poster2json/schemas/poster_schema.json +1181 -0
  10. poster2json-0.1.0/poster2json/standards.py +21 -0
  11. poster2json-0.1.0/poster2json/tests/__init__.py +1 -0
  12. poster2json-0.1.0/poster2json/tests/conftest.py +1 -0
  13. poster2json-0.1.0/poster2json/utils.py +172 -0
  14. poster2json-0.1.0/poster2json/validate.py +337 -0
  15. poster2json-0.1.0/pyproject.toml +143 -0
poster2json-0.1.0/LICENSE.md ADDED
@@ -0,0 +1,22 @@
1
+
2
+ MIT License
3
+
4
+ Copyright © 2022, FAIR Data Innovations Hub
5
+
6
+ Permission is hereby granted, free of charge, to any person obtaining a copy
7
+ of this software and associated documentation files (the "Software"), to deal
8
+ in the Software without restriction, including without limitation the rights
9
+ to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
10
+ copies of the Software, and to permit persons to whom the Software is
11
+ furnished to do so, subject to the following conditions:
12
+
13
+ The above copyright notice and this permission notice shall be included in all
14
+ copies or substantial portions of the Software.
15
+
16
+ THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
17
+ IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
18
+ FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
19
+ AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
20
+ LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
21
+ OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
22
+ SOFTWARE.
poster2json-0.1.0/PKG-INFO ADDED
@@ -0,0 +1,263 @@
1
+ Metadata-Version: 2.4
2
+ Name: poster2json
3
+ Version: 0.1.0
4
+ Summary: Convert scientific posters (PDF/images) to structured JSON metadata using Large Language Models
5
+ License: MIT
6
+ License-File: LICENSE.md
7
+ Keywords: poster,json,metadata,extraction,llm,scientific,pdf,ocr,machine-learning,datacite,fair-data
8
+ Author: FAIR Data Innovations Hub
9
+ Author-email: contact@fairdataihub.org
10
+ Requires-Python: >=3.10,<4.0
11
+ Classifier: Development Status :: 4 - Beta
12
+ Classifier: Intended Audience :: Developers
13
+ Classifier: Intended Audience :: Science/Research
14
+ Classifier: License :: OSI Approved :: MIT License
15
+ Classifier: Natural Language :: English
16
+ Classifier: Operating System :: OS Independent
17
+ Classifier: Programming Language :: Python
18
+ Classifier: Programming Language :: Python :: 3
19
+ Classifier: Programming Language :: Python :: 3.10
20
+ Classifier: Programming Language :: Python :: 3.11
21
+ Classifier: Programming Language :: Python :: 3.12
22
+ Classifier: Programming Language :: Python :: 3.13
23
+ Classifier: Programming Language :: Python :: 3.14
24
+ Classifier: Topic :: Scientific/Engineering
25
+ Classifier: Topic :: Text Processing
26
+ Provides-Extra: vision
27
+ Requires-Dist: Pillow (>=9.0.0)
28
+ Requires-Dist: accelerate (>=0.20.0)
29
+ Requires-Dist: art (>=6.0,<7.0)
30
+ Requires-Dist: bitsandbytes (>=0.44.0)
31
+ Requires-Dist: click (>=8.0,<9.0)
32
+ Requires-Dist: jsonschema (>=4.17.0)
33
+ Requires-Dist: numpy
34
+ Requires-Dist: pymupdf (>=1.22.0)
35
+ Requires-Dist: rouge-score
36
+ Requires-Dist: safetensors
37
+ Requires-Dist: sentencepiece
38
+ Requires-Dist: torch (>=2.0.0)
39
+ Requires-Dist: transformers (>=4.40.0)
40
+ Project-URL: Documentation, https://fairdataihub.github.io/poster2json
41
+ Project-URL: Homepage, https://github.com/fairdataihub/poster2json
42
+ Project-URL: Repository, https://github.com/fairdataihub/poster2json
43
+ Description-Content-Type: text/markdown
44
+
45
+ <div align="center">
46
+
47
+ <img src="https://raw.githubusercontent.com/fairdataihub/poster2json/main/logo.svg" alt="logo" width="200" height="auto" />
48
+
49
+ <br />
50
+
51
+ <h1>poster2json</h1>
52
+
53
+ <p>
54
+ Convert scientific posters (PDF/images) to structured JSON metadata using Large Language Models.
55
+ </p>
56
+
57
+ <br />
58
+
59
+ <p>
60
+ <a href="https://github.com/fairdataihub/poster2json/graphs/contributors">
61
+ <img src="https://img.shields.io/github/contributors/fairdataihub/poster2json.svg?style=flat-square" alt="contributors" />
62
+ </a>
63
+ <a href="https://github.com/fairdataihub/poster2json/stargazers">
64
+ <img src="https://img.shields.io/github/stars/fairdataihub/poster2json.svg?style=flat-square" alt="stars" />
65
+ </a>
66
+ <a href="https://github.com/fairdataihub/poster2json/issues/">
67
+ <img src="https://img.shields.io/github/issues/fairdataihub/poster2json.svg?style=flat-square" alt="open issues" />
68
+ </a>
69
+ <a href="https://github.com/fairdataihub/poster2json/blob/main/LICENSE">
70
+ <img src="https://img.shields.io/github/license/fairdataihub/poster2json.svg?style=flat-square" alt="license" />
71
+ </a>
72
+ </p>
73
+ <p>
74
+ <a href="https://pypi.org/project/poster2json">
75
+ <img src="https://img.shields.io/pypi/v/poster2json.svg" alt="PyPI Version" />
76
+ </a>
77
+ <a href="https://pypistats.org/packages/poster2json">
78
+ <img src="https://img.shields.io/pypi/dm/poster2json.svg?color=orange" alt="PyPI Downloads" />
79
+ </a>
80
+ <a href="https://zenodo.org/badge/latestdoi/1105067405">
81
+ <img src="https://zenodo.org/badge/1105067405.svg" alt="DOI" />
82
+ </a>
83
+ </p>
84
+
85
+ <h4>
86
+ <a href="https://fairdataihub.github.io/poster2json/">Documentation</a>
87
+ <span> · </span>
88
+ <a href="https://fairdataihub.github.io/poster2json/about/changelog/">Changelog</a>
89
+ <span> · </span>
90
+ <a href="https://github.com/fairdataihub/poster2json/issues/">Report Bug</a>
91
+ <span> · </span>
92
+ <a href="https://github.com/fairdataihub/poster2json/issues/">Request Feature</a>
93
+ </h4>
94
+ </div>
95
+
96
+ <br />
97
+
98
+ ---
99
+
100
+ ## Description
101
+
102
+ **poster2json** extracts structured metadata from scientific conference posters (PDF or image format) into machine-actionable JSON conforming to the [poster-json-schema](https://github.com/fairdataihub/poster-json-schema).
103
+
104
+ The pipeline uses:
105
+
106
+ - **Llama 3.1 8B** (fine-tuned) for JSON structuring
107
+ - **Qwen2-VL-7B** for vision-based OCR of image posters
108
+ - **pdfalto** for layout-aware PDF text extraction
109
+
110
+ ## Quick Start
111
+
112
+ ### Installation
113
+
114
+ ```bash
115
+ pip install poster2json
116
+ ```
117
+
118
+ ### CLI Usage
119
+
120
+ ```bash
121
+ # Extract metadata from a poster
122
+ poster2json extract poster.pdf -o result.json
123
+
124
+ # Validate extracted JSON
125
+ poster2json validate result.json
126
+
127
+ # Process multiple posters
128
+ poster2json batch ./posters/ -o ./output/
129
+ ```
130
+
131
+ ### Python API
132
+
133
+ ```python
134
+ from poster2json import extract_poster, validate_poster
135
+
136
+ # Extract metadata
137
+ result = extract_poster("poster.pdf")
138
+ print(result["titles"][0]["title"])
139
+
140
+ # Validate the result
141
+ is_valid = validate_poster(result)
142
+ ```
143
+
144
+ ## Output Format
145
+
146
+ Output conforms to the [poster-json-schema](https://github.com/fairdataihub/poster-json-schema) (DataCite-based):
147
+
148
+ ```json
149
+ {
150
+ "$schema": "https://posters.science/schema/v0.1/poster_schema.json",
151
+ "creators": [
152
+ {
153
+ "name": "Garcia, Sofia",
154
+ "givenName": "Sofia",
155
+ "familyName": "Garcia",
156
+ "affiliation": ["University"]
157
+ }
158
+ ],
159
+ "titles": [
160
+ { "title": "Machine Learning Approaches to Diabetic Retinopathy Detection" }
161
+ ],
162
+ "posterContent": {
163
+ "sections": [
164
+ { "sectionTitle": "Abstract", "sectionContent": "..." },
165
+ { "sectionTitle": "Methods", "sectionContent": "..." },
166
+ { "sectionTitle": "Results", "sectionContent": "..." }
167
+ ]
168
+ },
169
+ "imageCaptions": [{ "captions": ["Figure 1.", "ROC curves showing..."] }],
170
+ "tableCaptions": [{ "captions": ["Table 1.", "Performance metrics"] }]
171
+ }
172
+ ```
173
+
174
+ ## System Requirements
175
+
176
+ | Requirement | Specification |
177
+ | ----------- | -------------------------------- |
178
+ | GPU | NVIDIA CUDA-capable, ≥16GB VRAM |
179
+ | RAM | ≥32GB recommended |
180
+ | Python | 3.10+ |
181
+ | OS | Linux, macOS, Windows (via WSL2) |
182
+
183
+ ## Performance
184
+
185
+ Validated on 10 manually annotated scientific posters:
186
+
187
+ | Metric | Score | Threshold |
188
+ | ---------------- | ----- | --------- |
189
+ | Word Capture | 0.96 | ≥0.75 |
190
+ | ROUGE-L | 0.89 | ≥0.75 |
191
+ | Number Capture | 0.93 | ≥0.75 |
192
+ | Field Proportion | 0.99 | 0.30–2.50 |
193
+
194
+ **Pass Rate**: 10/10 (100%)
195
+
196
+ ## Documentation
197
+
198
+ | Document | Description |
199
+ | ------------------------------------ | ------------------------------- |
200
+ | [Architecture](docs/architecture.md) | Technical details & methodology |
201
+ | [Evaluation](docs/evaluation.md) | Validation metrics & results |
202
+
203
+ ## Development Setup
204
+
205
+ ```bash
206
+ # Clone the repository
207
+ git clone https://github.com/fairdataihub/poster2json.git
208
+ cd poster2json
209
+
210
+ # Create a virtual environment
211
+ python -m venv .venv
212
+
213
+ # Activate the virtual environment
214
+ source venv/bin/activate # On Windows: .venv\Scripts\activate
215
+
216
+ # Install poetry
217
+ pip install poetry
218
+
219
+ # Install dependencies
220
+ poetry install
221
+
222
+ # Run tests
223
+ poe test
224
+
225
+ # Format code
226
+ poe format
227
+ ```
228
+
229
+ If you are on windows and have multiple python versions, you can use the following commands:
230
+
231
+ ```bash
232
+ py -0p # list all python versions
233
+
234
+ py -3.12 -m venv .venv
235
+ ```
236
+
237
+ ## License
238
+
239
+ MIT License - see [LICENSE](LICENSE.md) for details.
240
+
241
+ ## Citation
242
+
243
+ ```bibtex
244
+ @software{poster2json2026,
245
+ title = {poster2json: Scientific Poster to JSON Metadata Extraction},
246
+ author = {O'Neill, James and Soundarajan, Sanjay and Portillo, Dorian and Patel, Bhavesh},
247
+ year = {2026},
248
+ url = {https://github.com/fairdataihub/poster2json},
249
+ doi = {10.5281/zenodo.18320010}
250
+ }
251
+ ```
252
+
253
+ ## Acknowledgements
254
+
255
+ - [FAIR Data Innovations Hub](https://fairdataihub.org/)
256
+ - Meta AI for Llama 3.1
257
+ - Alibaba Cloud for Qwen2-VL
258
+ - Part of the [posters.science](https://posters.science) platform
259
+
260
+ ## Contributing
261
+
262
+ Contributions welcome! Please see [CONTRIBUTING.md](CONTRIBUTING.md) for guidelines.
263
+
poster2json-0.1.0/README.md ADDED
@@ -0,0 +1,218 @@
1
+ <div align="center">
2
+
3
+ <img src="https://raw.githubusercontent.com/fairdataihub/poster2json/main/logo.svg" alt="logo" width="200" height="auto" />
4
+
5
+ <br />
6
+
7
+ <h1>poster2json</h1>
8
+
9
+ <p>
10
+ Convert scientific posters (PDF/images) to structured JSON metadata using Large Language Models.
11
+ </p>
12
+
13
+ <br />
14
+
15
+ <p>
16
+ <a href="https://github.com/fairdataihub/poster2json/graphs/contributors">
17
+ <img src="https://img.shields.io/github/contributors/fairdataihub/poster2json.svg?style=flat-square" alt="contributors" />
18
+ </a>
19
+ <a href="https://github.com/fairdataihub/poster2json/stargazers">
20
+ <img src="https://img.shields.io/github/stars/fairdataihub/poster2json.svg?style=flat-square" alt="stars" />
21
+ </a>
22
+ <a href="https://github.com/fairdataihub/poster2json/issues/">
23
+ <img src="https://img.shields.io/github/issues/fairdataihub/poster2json.svg?style=flat-square" alt="open issues" />
24
+ </a>
25
+ <a href="https://github.com/fairdataihub/poster2json/blob/main/LICENSE">
26
+ <img src="https://img.shields.io/github/license/fairdataihub/poster2json.svg?style=flat-square" alt="license" />
27
+ </a>
28
+ </p>
29
+ <p>
30
+ <a href="https://pypi.org/project/poster2json">
31
+ <img src="https://img.shields.io/pypi/v/poster2json.svg" alt="PyPI Version" />
32
+ </a>
33
+ <a href="https://pypistats.org/packages/poster2json">
34
+ <img src="https://img.shields.io/pypi/dm/poster2json.svg?color=orange" alt="PyPI Downloads" />
35
+ </a>
36
+ <a href="https://zenodo.org/badge/latestdoi/1105067405">
37
+ <img src="https://zenodo.org/badge/1105067405.svg" alt="DOI" />
38
+ </a>
39
+ </p>
40
+
41
+ <h4>
42
+ <a href="https://fairdataihub.github.io/poster2json/">Documentation</a>
43
+ <span> · </span>
44
+ <a href="https://fairdataihub.github.io/poster2json/about/changelog/">Changelog</a>
45
+ <span> · </span>
46
+ <a href="https://github.com/fairdataihub/poster2json/issues/">Report Bug</a>
47
+ <span> · </span>
48
+ <a href="https://github.com/fairdataihub/poster2json/issues/">Request Feature</a>
49
+ </h4>
50
+ </div>
51
+
52
+ <br />
53
+
54
+ ---
55
+
56
+ ## Description
57
+
58
+ **poster2json** extracts structured metadata from scientific conference posters (PDF or image format) into machine-actionable JSON conforming to the [poster-json-schema](https://github.com/fairdataihub/poster-json-schema).
59
+
60
+ The pipeline uses:
61
+
62
+ - **Llama 3.1 8B** (fine-tuned) for JSON structuring
63
+ - **Qwen2-VL-7B** for vision-based OCR of image posters
64
+ - **pdfalto** for layout-aware PDF text extraction
65
+
66
+ ## Quick Start
67
+
68
+ ### Installation
69
+
70
+ ```bash
71
+ pip install poster2json
72
+ ```
73
+
74
+ ### CLI Usage
75
+
76
+ ```bash
77
+ # Extract metadata from a poster
78
+ poster2json extract poster.pdf -o result.json
79
+
80
+ # Validate extracted JSON
81
+ poster2json validate result.json
82
+
83
+ # Process multiple posters
84
+ poster2json batch ./posters/ -o ./output/
85
+ ```
86
+
87
+ ### Python API
88
+
89
+ ```python
90
+ from poster2json import extract_poster, validate_poster
91
+
92
+ # Extract metadata
93
+ result = extract_poster("poster.pdf")
94
+ print(result["titles"][0]["title"])
95
+
96
+ # Validate the result
97
+ is_valid = validate_poster(result)
98
+ ```
99
+
100
+ ## Output Format
101
+
102
+ Output conforms to the [poster-json-schema](https://github.com/fairdataihub/poster-json-schema) (DataCite-based):
103
+
104
+ ```json
105
+ {
106
+ "$schema": "https://posters.science/schema/v0.1/poster_schema.json",
107
+ "creators": [
108
+ {
109
+ "name": "Garcia, Sofia",
110
+ "givenName": "Sofia",
111
+ "familyName": "Garcia",
112
+ "affiliation": ["University"]
113
+ }
114
+ ],
115
+ "titles": [
116
+ { "title": "Machine Learning Approaches to Diabetic Retinopathy Detection" }
117
+ ],
118
+ "posterContent": {
119
+ "sections": [
120
+ { "sectionTitle": "Abstract", "sectionContent": "..." },
121
+ { "sectionTitle": "Methods", "sectionContent": "..." },
122
+ { "sectionTitle": "Results", "sectionContent": "..." }
123
+ ]
124
+ },
125
+ "imageCaptions": [{ "captions": ["Figure 1.", "ROC curves showing..."] }],
126
+ "tableCaptions": [{ "captions": ["Table 1.", "Performance metrics"] }]
127
+ }
128
+ ```
129
+
130
+ ## System Requirements
131
+
132
+ | Requirement | Specification |
133
+ | ----------- | -------------------------------- |
134
+ | GPU | NVIDIA CUDA-capable, ≥16GB VRAM |
135
+ | RAM | ≥32GB recommended |
136
+ | Python | 3.10+ |
137
+ | OS | Linux, macOS, Windows (via WSL2) |
138
+
139
+ ## Performance
140
+
141
+ Validated on 10 manually annotated scientific posters:
142
+
143
+ | Metric | Score | Threshold |
144
+ | ---------------- | ----- | --------- |
145
+ | Word Capture | 0.96 | ≥0.75 |
146
+ | ROUGE-L | 0.89 | ≥0.75 |
147
+ | Number Capture | 0.93 | ≥0.75 |
148
+ | Field Proportion | 0.99 | 0.30–2.50 |
149
+
150
+ **Pass Rate**: 10/10 (100%)
151
+
152
+ ## Documentation
153
+
154
+ | Document | Description |
155
+ | ------------------------------------ | ------------------------------- |
156
+ | [Architecture](docs/architecture.md) | Technical details & methodology |
157
+ | [Evaluation](docs/evaluation.md) | Validation metrics & results |
158
+
159
+ ## Development Setup
160
+
161
+ ```bash
162
+ # Clone the repository
163
+ git clone https://github.com/fairdataihub/poster2json.git
164
+ cd poster2json
165
+
166
+ # Create a virtual environment
167
+ python -m venv .venv
168
+
169
+ # Activate the virtual environment
170
+ source venv/bin/activate # On Windows: .venv\Scripts\activate
171
+
172
+ # Install poetry
173
+ pip install poetry
174
+
175
+ # Install dependencies
176
+ poetry install
177
+
178
+ # Run tests
179
+ poe test
180
+
181
+ # Format code
182
+ poe format
183
+ ```
184
+
185
+ If you are on windows and have multiple python versions, you can use the following commands:
186
+
187
+ ```bash
188
+ py -0p # list all python versions
189
+
190
+ py -3.12 -m venv .venv
191
+ ```
192
+
193
+ ## License
194
+
195
+ MIT License - see [LICENSE](LICENSE.md) for details.
196
+
197
+ ## Citation
198
+
199
+ ```bibtex
200
+ @software{poster2json2026,
201
+ title = {poster2json: Scientific Poster to JSON Metadata Extraction},
202
+ author = {O'Neill, James and Soundarajan, Sanjay and Portillo, Dorian and Patel, Bhavesh},
203
+ year = {2026},
204
+ url = {https://github.com/fairdataihub/poster2json},
205
+ doi = {10.5281/zenodo.18320010}
206
+ }
207
+ ```
208
+
209
+ ## Acknowledgements
210
+
211
+ - [FAIR Data Innovations Hub](https://fairdataihub.org/)
212
+ - Meta AI for Llama 3.1
213
+ - Alibaba Cloud for Qwen2-VL
214
+ - Part of the [posters.science](https://posters.science) platform
215
+
216
+ ## Contributing
217
+
218
+ Contributions welcome! Please see [CONTRIBUTING.md](CONTRIBUTING.md) for guidelines.
poster2json-0.1.0/poster2json/__init__.py ADDED
@@ -0,0 +1,47 @@
1
+ """
2
+ poster2json - Convert scientific posters to structured JSON metadata.
3
+
4
+ Extract structured metadata from scientific poster PDFs and images
5
+ using Large Language Models. Output conforms to the poster-json-schema
6
+ (DataCite-based format).
7
+
8
+ Basic Usage:
9
+ >>> from poster2json import extract_poster, validate_poster
10
+ >>>
11
+ >>> # Extract metadata from a poster
12
+ >>> result = extract_poster("poster.pdf")
13
+ >>> print(result["titles"][0]["title"])
14
+
15
+ >>> # Validate extracted JSON
16
+ >>> is_valid = validate_poster(result)
17
+
18
+ CLI Usage:
19
+ $ poster2json extract poster.pdf -o result.json
20
+ $ poster2json validate result.json
21
+ """
22
+
23
+ from importlib.metadata import PackageNotFoundError, version
24
+
25
+ # Main functions
26
+ from .extract import extract_poster
27
+ from .validate import (
28
+ get_validation_errors,
29
+ validate_comprehensive,
30
+ validate_poster,
31
+ )
32
+
33
+ try:
34
+ __version__ = version("poster2json")
35
+ except PackageNotFoundError:
36
+ __version__ = "(local)"
37
+
38
+ del PackageNotFoundError
39
+ del version
40
+
41
+ __all__ = [
42
+ "extract_poster",
43
+ "validate_poster",
44
+ "validate_comprehensive",
45
+ "get_validation_errors",
46
+ "__version__",
47
+ ]
poster2json-0.1.0/poster2json/__main__.py ADDED
@@ -0,0 +1,8 @@
1
+ #!/usr/bin/env python
2
+
3
+ """Package entry point."""
4
+
5
+ from poster2json.cli import main
6
+
7
+ if __name__ == "__main__": # pragma: no cover
8
+ main() # pylint: disable=no-value-for-parameter