Marina Oliveira
Papers
4
Total Citations
250
H-Index
3
About
Marina Oliveira is a leading researcher at the intersection of computer vision and medical robotics, with a primary focus on advancing endoscopic technologies. Her most impactful contribution is the creation of the **EndoSLAM dataset**, a comprehensive benchmark that has become essential for quantitatively evaluating simultaneous localization and mapping (SLAM) and depth estimation methods in endoscopic videos. This work, published in 2021, has garnered **240 citations**, reflecting its critical role in filling a gap where no effective benchmarking tools previously existed. Oliveira’s key research areas include **monocular visual odometry**, **unsupervised depth estimation**, and **robotic validation for capsule endoscopy**. Notably, she developed Endo-SfMLearner, an unsupervised deep learning approach that enables dense topography reconstruction and pose estimation without requiring ground-truth labels. Her earlier work on wireless capsule endoscope location, validated through robotic experiments, further underscores her commitment to translating algorithmic advances into practical medical tools. Through these contributions, Oliveira has provided the research community with foundational resources and methods that are driving progress in minimally invasive diagnostics and surgical navigation.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3Quantitative Evaluation of Endoscopic SLAM Methods: EndoSLAM Dataset.3 citations · 2020
- 4Wireless Capsule Endoscope Location and a Robotic Validation Experiment2 citations · 2019