Dataset Open Access
Metric-Guided Synthetic Image Data Rendering for Deep Learning compatible with Agentic AI
Radoynova, Martina;
Pantze, Samuel;
De, Trina;
Günther, Ulrik;
Yakimovich, Artur
MARC21 XML Export
<?xml version='1.0' encoding='UTF-8'?>
<record xmlns="http://www.loc.gov/MARC21/slim">
<leader>00000nmm##2200000uu#4500</leader>
<controlfield tag="001">4780</controlfield>
<datafield tag="773" ind1=" " ind2=" ">
<subfield code="a">https://www.hzdr.de/publications/Publ-43634</subfield>
<subfield code="i">isIdenticalTo</subfield>
<subfield code="n">url</subfield>
</datafield>
<datafield tag="773" ind1=" " ind2=" ">
<subfield code="a">10.14278/rodare.4779</subfield>
<subfield code="i">isVersionOf</subfield>
<subfield code="n">doi</subfield>
</datafield>
<datafield tag="100" ind1=" " ind2=" ">
<subfield code="a">Radoynova, Martina</subfield>
<subfield code="u">Center for Advanced Systems Understanding (CASUS), Görlitz, Germany</subfield>
<subfield code="0">(orcid)0009-0003-4358-0391</subfield>
</datafield>
<datafield tag="980" ind1=" " ind2=" ">
<subfield code="a">dataset</subfield>
</datafield>
<datafield tag="650" ind1="1" ind2="7">
<subfield code="a">cc-by</subfield>
<subfield code="2">opendefinition.org</subfield>
</datafield>
<datafield tag="856" ind1="4" ind2=" ">
<subfield code="s">253530792</subfield>
<subfield code="u">https://rodare.hzdr.de/record/4780/files/blender.zip</subfield>
<subfield code="z">md5:b7b9b33cbe3f1a233c4be98f0fcb6d90</subfield>
</datafield>
<datafield tag="856" ind1="4" ind2=" ">
<subfield code="s">798358357</subfield>
<subfield code="u">https://rodare.hzdr.de/record/4780/files/dataset_high-L.zip</subfield>
<subfield code="z">md5:3b0ee6b6a2933b7e508ddbd9acb5a52f</subfield>
</datafield>
<datafield tag="856" ind1="4" ind2=" ">
<subfield code="s">79701167</subfield>
<subfield code="u">https://rodare.hzdr.de/record/4780/files/dataset_high-S.zip</subfield>
<subfield code="z">md5:2152030f1bd2fa389b093def30d2c8cf</subfield>
</datafield>
<datafield tag="856" ind1="4" ind2=" ">
<subfield code="s">527874457</subfield>
<subfield code="u">https://rodare.hzdr.de/record/4780/files/dataset_low-L.zip</subfield>
<subfield code="z">md5:25f72dac4902d12fcd8b5d4b29213153</subfield>
</datafield>
<datafield tag="856" ind1="4" ind2=" ">
<subfield code="s">52122808</subfield>
<subfield code="u">https://rodare.hzdr.de/record/4780/files/dataset_low-S.zip</subfield>
<subfield code="z">md5:67df196138195355800e6bbcc4324307</subfield>
</datafield>
<datafield tag="856" ind1="4" ind2=" ">
<subfield code="s">616510707</subfield>
<subfield code="u">https://rodare.hzdr.de/record/4780/files/dataset_medium-L.zip</subfield>
<subfield code="z">md5:0798aa898c0757ae37016e07b3cf99f1</subfield>
</datafield>
<datafield tag="856" ind1="4" ind2=" ">
<subfield code="s">61594229</subfield>
<subfield code="u">https://rodare.hzdr.de/record/4780/files/dataset_medium-S.zip</subfield>
<subfield code="z">md5:06f4c8e9250d01ccf76e26cf228d428f</subfield>
</datafield>
<datafield tag="856" ind1="4" ind2=" ">
<subfield code="s">78937319</subfield>
<subfield code="u">https://rodare.hzdr.de/record/4780/files/dataset_mix.zip</subfield>
<subfield code="z">md5:62f86b529ba7f985b6f4439dd4d1ea42</subfield>
</datafield>
<datafield tag="856" ind1="4" ind2=" ">
<subfield code="s">223507341</subfield>
<subfield code="u">https://rodare.hzdr.de/record/4780/files/model_high-L.zip</subfield>
<subfield code="z">md5:d36bf6e93956cea3e04babe2793498ff</subfield>
</datafield>
<datafield tag="856" ind1="4" ind2=" ">
<subfield code="s">222653860</subfield>
<subfield code="u">https://rodare.hzdr.de/record/4780/files/model_high-S.zip</subfield>
<subfield code="z">md5:a1d568aca0dc863a4e897ee5a11c1f61</subfield>
</datafield>
<datafield tag="856" ind1="4" ind2=" ">
<subfield code="s">223659983</subfield>
<subfield code="u">https://rodare.hzdr.de/record/4780/files/model_low-L.zip</subfield>
<subfield code="z">md5:d2748250e5704cd9a884fbd593318edf</subfield>
</datafield>
<datafield tag="520" ind1=" " ind2=" ">
<subfield code="a"><p>This repository contains the datasets, 3D rendering scene files, and trained model weights accompanying the manuscript <strong>&quot;Metric-Guided Synthetic Image Data Rendering for Deep Learning compatible with Agentic AI.&quot;</strong> The deposited files provide the complete procedural pipeline and resulting artefacts used to evaluate zero-shot instance segmentation of virological plaque assays (based on the <a href="https://rodare.hzdr.de/record/3003">VACVPlaque dataset</a>).</p>
<p>Repository Contents</p>
<ul>
<li>
<p><strong><code>datasets/</code></strong> Contains the synthetic image datasets generated at three varying levels of perceived realism (High, Medium, Low) and in two sizes (Small [S]: 100 images; Large [L]: 1,000 images). It also includes the <code>Mix</code> dataset, which consists of 90 high-realism synthetic images and 10 real images.</p>
</li>
<li>
<p><strong><code>blender/</code></strong> Includes the procedural 3D scene files used to generate the synthetic datasets via Blender. These files can be used manually or coupled with the SynthClaw agentic skill for automated rendering.</p>
<ul>
<li>
<p><code>Scene_High.blend</code></p>
</li>
<li>
<p><code>Scene_Medium.blend</code></p>
</li>
<li>
<p><code>Scene_Low.blend</code></p>
</li>
</ul>
</li>
<li>
<p><strong><code>model_weights/</code></strong> Contains the trained single-shot architecture model weights and threshold configurations used to evaluate the zero-shot performance of each generated dataset.</p>
<ul>
<li>
<p><strong>Evaluated Conditions:</strong> <code>High_L</code>, <code>High_S</code>, <code>Medium_L</code>, <code>Medium_S</code>, <code>Low_L</code>, <code>Low_S</code>, and <code>Mix</code>.</p>
</li>
<li>
<p><strong>Files per Condition:</strong> Each subdirectory includes the best and last model weights (<code>weights_best.h5</code>, <code>weights_last.h5</code>) alongside the corresponding threshold optimization parameters (<code>thresholds1.json</code>, <code>thresholds2.json</code>) for large and small object segmentation.</p>
</li>
</ul>
</li>
</ul></subfield>
</datafield>
<datafield tag="024" ind1=" " ind2=" ">
<subfield code="a">10.14278/rodare.4780</subfield>
<subfield code="2">doi</subfield>
</datafield>
<datafield tag="542" ind1=" " ind2=" ">
<subfield code="l">open</subfield>
</datafield>
<datafield tag="980" ind1=" " ind2=" ">
<subfield code="a">user-health</subfield>
</datafield>
<datafield tag="980" ind1=" " ind2=" ">
<subfield code="a">user-rodare</subfield>
</datafield>
<datafield tag="540" ind1=" " ind2=" ">
<subfield code="u">https://creativecommons.org/licenses/by/4.0/legalcode</subfield>
<subfield code="a">Creative Commons Attribution 4.0 International</subfield>
</datafield>
<datafield tag="653" ind1=" " ind2=" ">
<subfield code="a">deep learning</subfield>
</datafield>
<datafield tag="653" ind1=" " ind2=" ">
<subfield code="a">synthetic data</subfield>
</datafield>
<datafield tag="653" ind1=" " ind2=" ">
<subfield code="a">procedural image rendering</subfield>
</datafield>
<datafield tag="653" ind1=" " ind2=" ">
<subfield code="a">Blender</subfield>
</datafield>
<datafield tag="653" ind1=" " ind2=" ">
<subfield code="a">computer vision</subfield>
</datafield>
<datafield tag="653" ind1=" " ind2=" ">
<subfield code="a">instance segmentation</subfield>
</datafield>
<datafield tag="700" ind1=" " ind2=" ">
<subfield code="a">Pantze, Samuel</subfield>
<subfield code="u">Center for Advanced Systems Understanding (CASUS), Görlitz, Germany</subfield>
<subfield code="0">(orcid)0009-0000-1388-8959</subfield>
</datafield>
<datafield tag="700" ind1=" " ind2=" ">
<subfield code="a">De, Trina</subfield>
<subfield code="u">Center for Advanced Systems Understanding (CASUS), Görlitz, Germany</subfield>
<subfield code="0">(orcid)0000-0003-1111-9851</subfield>
</datafield>
<datafield tag="700" ind1=" " ind2=" ">
<subfield code="a">Günther, Ulrik</subfield>
<subfield code="u">Center for Advanced Systems Understanding (CASUS), Görlitz, Germany</subfield>
<subfield code="0">(orcid)0000-0002-1179-8228</subfield>
</datafield>
<datafield tag="700" ind1=" " ind2=" ">
<subfield code="a">Yakimovich, Artur</subfield>
<subfield code="u">Center for Advanced Systems Understanding (CASUS), Görlitz, Germany</subfield>
<subfield code="0">(orcid)0000-0003-2458-4904</subfield>
</datafield>
<datafield tag="909" ind1="C" ind2="O">
<subfield code="o">oai:rodare.hzdr.de:4780</subfield>
<subfield code="p">openaire_data</subfield>
<subfield code="p">user-health</subfield>
<subfield code="p">user-rodare</subfield>
</datafield>
<datafield tag="245" ind1=" " ind2=" ">
<subfield code="a">Metric-Guided Synthetic Image Data Rendering for Deep Learning compatible with Agentic AI</subfield>
</datafield>
<controlfield tag="005">20260713100227.0</controlfield>
<datafield tag="260" ind1=" " ind2=" ">
<subfield code="c">2026-07-13</subfield>
</datafield>
<datafield tag="041" ind1=" " ind2=" ">
<subfield code="a">eng</subfield>
</datafield>
</record>
34
0
views
downloads
| All versions | This version | |
|---|---|---|
| Views | 34 | 34 |
| Downloads | 0 | 0 |
| Data volume | 0 Bytes | 0 Bytes |
| Unique views | 30 | 30 |
| Unique downloads | 0 | 0 |
- Publication date:
- July 13, 2026
- DOI:
-
- Keyword(s):
- deep learning synthetic data procedural image rendering Blender computer vision instance segmentation
- Related identifiers:
- Identical to:
https://www.hzdr.de/publications/Publ-43634 - Communities:
- License (for files):
- Creative Commons Attribution 4.0 International