Juana Valeria Hurtado
Papers
7
Total Citations
265
H-Index
4
About
Juana Valeria Hurtado is a leading researcher at the intersection of robot perception, scene understanding, and ethical AI. Her primary contributions lie in developing comprehensive frameworks for how robots and autonomous vehicles perceive and navigate dynamic urban environments. She is best known for her pioneering work on panoptic scene understanding, where she co-created the **Panoptic nuScenes** benchmark (183 citations), a large-scale LiDAR dataset that has become a standard for evaluating panoptic segmentation and tracking. Her **MOPT** (Multi-Object Panoptic Tracking) framework further advanced this field by unifying detection, segmentation, and tracking into a single model, enabling robots to achieve a holistic understanding of moving agents and static objects simultaneously. Beyond perception, Hurtado is a vocal advocate for fairness in robotics. Her recent work on **Fairness and Bias in Robot Learning** (2024) critically examines how machine learning models can replicate societal discrimination, and she has introduced feminist perspectives to robot learning processes. By bridging technical innovation with social responsibility, Hurtado is shaping a future where intelligent systems are not only more perceptive but also more equitable.
Research Focus
Key Achievements
Top Papers
- 1
- 2Semantic scene segmentation for robotics44 citations · 2022
- 3MOPT: Multi-Object Panoptic Tracking21 citations · 2020
- 4Fairness and Bias in Robot Learning9 citations · 2024
- 5Fairness and Bias in Robot Learning4 citations · 2022
- 6Feminist Perspective on Robot Learning Processes2 citations · 2022
- 7