[visionlist] [News] Release of three surgical datasets for scene segmentation, object detection, VQA and activity recognition

[Nicolas Padoy]

Dear all,

We are pleased to announce the release of three richly annotated 
datasets for surgical video analysis:

1. Endoscapes is a dataset of 201 laparoscopic videos richly annotated 
for scene segmentation, object detection, and critical view of safety 
assessment.
    arXiv: https://arxiv.org/abs/2312.12429
    code: https://github.com/CAMMA-public/Endoscapes

2. SSG-VQA is a Visual Question Answering dataset on laparoscopic videos 
providing diverse, geometrically grounded, unbiased, and surgical 
action-oriented queries generated using surgical scene graphs.
    arXiv: https://arxiv.org/abs/2312.10251
code: https://github.com/CAMMA-public/SSG-VQA

3. Multibypass140 is a multicentric dataset of endoscopic video of 
laparoscopic Roux-en-Y gastric bypass surgery for research on 
multi-level surgical activity recognition, specifically phases, and steps.
    arXiv: https://arxiv.org/abs/2312.11250
code: https://github.com/CAMMA-public/MultiBypass140


Kind regards,

Nicolas Padoy

-- 
Nicolas Padoy
Professor of Computer Science, University of Strasbourg
Scientific Director, IHU Strasbourg
Director of CS & AI Research, IHU Strasbourg
Head of Research Group CAMMA
Web: http://camma.u-strasbg.fr

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://visionscience.com/pipermail/visionlist_visionscience.com/attachments/20240115/7f8d77af/attachment-0001.html>