DREAMING - Diminished Reality for Emerging Applications in Medicine through Inpainting Dataset

<p>Dataset for the <strong>DREAMING - Diminished Reality for Emerging Applications in Medicine through Inpainting Challenge</strong></p> <p>More information about the challenge and challenge registration can be found on the <a href="https://dreaming.grand-challenge.org/" target="_blank" rel="noopener">challenge platform</a>!</p> <p>&nbsp;</p> <p><em><strong>Timeline:</strong></em></p> <p><strong>8th January 2024: </strong>First subset of training &amp; validation data available (scenes 0000-0009).</p> <p><strong>22nd January 2024:</strong> Second subset of training &amp; validation data available (scenes 0010-0049).</p> <p><strong>26th January 2024:</strong> Full training &amp; validation data available (scenes 0050-0100).</p> <p>&nbsp;</p> <p><strong>Description:</strong></p> <p>The dataset was created using&nbsp;<a href="https://www.unrealengine.com">Unreal Engine 5.1</a>, <a href="https://www.unrealengine.com/en-US/metahuman">Unreal MetaHumans</a>, <a href="../records/10091715" target="_blank" rel="noopener">3D-COSI surgical instruments</a>, <a href="https://github.com/BatFaceWayne/POV_Surgery" target="_blank" rel="noopener">POV-Surgery grasp generation</a> and <a href="https://github.com/ydrive/EasySynth" target="_blank" rel="noopener">EasySynth</a>.</p> <p>Each scene contains:</p> <ul> <li>"color": RGB input images</li> <li>"gt": RGB ground truth images</li> <li>"mask": Mask defining the area to be inpainted. White -&gt; Background, Black -&gt; Inpainting area</li> <li>"CameraPoses.csv": Camera poses</li> <li>"CameraRig.json": Camera intrinsics</li> </ul> <p>To cite the dataset, please use</p> <pre><code>@online{gsaxner_2024_10570773, author = {Gsaxner, Christina and van Meegdenburg, Timo and Luijten, Gijs and Vu, Viet Duc and Puladi, Behrus and Egger, Jan}, title = {{DREAMING - Diminished Reality for Emerging Applications in Medicine through Inpainting Dataset}}, month = jan, year = 2024, publisher = {Zenodo}, doi = {10.5281/zenodo.10570773}, url = {https://doi.org/10.5281/zenodo.10570773} }</code></pre> <p>&nbsp;</p>