Federica Spinola
PhD Candidate in the Willow team (INRIA Paris), Msc. in Robotics at ETH Zurich
PhD Candidate in the Willow team (INRIA Paris), Msc. in Robotics at ETH Zurich
I am currently a first year PhD student working in the Willow team at INRIA Paris under the supervision of Prof. Cordelia Schmid. As an ELLIS PhD student, I am also co-supervised by Prof. Robert Katzschmann at ETH Zurich. My interests lie in leveraging human videos to teach robots dexterous manipulation tasks, improving generality, cross-embodiment transfer, and minimizing data collection efforts.
Previously, I worked as a Machine Learning Engineer at Deeping Source, a startup in South Korea focused on video understanding for retail. As a member of the research team, I engaged both in research and development under the supervision of Dr. Philipp Benz. My main research focused on Multi-Target Multi-Camera tracking (MTMC) for pedestrian journey analysis. I primarily tackled problems related to re-identification, learning from poorly labeled data, and data generation/auto-labeling. I also supported research in model quantization, action recognition and monocular camera calibration.
Before this, I completed my MSc. in Robotics Systems and Control at ETH Zurich. Throughout my Master's I was supervised by Prof. Dr. Roland Siegwart. For my Master's thesis I worked on data driven methods for camera calibration of soccer broadcasting cameras at the Advanced Interactive Technologies (AIT) Lab under the supervision of Dr. Jie Song, Dr. Martin Oswald and Dr. Viktor Larsson.
Prior to this, I received a MEng in Mechanical Engineering from Imperial College London, where I learned the basics of control and mechanical design.
Oct 1, 2025
I just started my PhD in the Willow team of INRIA Paris under the supervision of Prof. Cordelia Schmid. I will be working on VLAs, with a focus on dexterous manipulation and learning from human videos.
Sep 3, 2024
I am applying to PhDs in Robotics. My research interests are related to the data challenge in robotic learning. I am interested in exploring ways to leverage simulation and existing data, and to minimize and automate data collection efforts to support research and deployment of robots in the real world. Feel free to contact me if you have open positions or if you want to discuss further.
Aug 2, 2024
Our paper "RoomRecon: High-Quality Textured Room Layout Reconstruction on Mobile Devices" got accepted to ISMAR 2024 and got nomitated for the Best Paper Award.
Oct 20, 2023
Our paper "A*: Atrous spatial temporal action recognition for real time applications" has been accepted to WACV 2024. Feel free to come and check it out!
Apr 3, 2023
Our paper "Knowledge Assembly: Semi-Supervised Multi-Task Learning from Multiple Datasets with Disjoint Labels" got accepted to the L3D-IVU workshop at CVPR 2023. Come by our poster if you are interested!
Jul 1, 2022
I am excited to start a new adventure in South Korea. I will be joining Deeping Source, an AI startup focused on video analytics and anonymization, as a Machine Learning Engineer.
RoomRecon: High-Quality Textured Room Layout Reconstruction on Mobile Devices
Kim S, Spinola F, Duc Cd, Lee S, Cho K.
In Proceedings of the IEEE International Symposium on Mixed and Augmented Reality. 2024.
A*: Atrous spatial temporal action recognition for real time applications
Kim M, Spinola F, Benz P, Kim TH.
In Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision. 2024.
Knowledge Assembly: Semi-Supervised Multi-Task Learning from Multiple Datasets with Disjoint Labels
Spinola F, Benz P, Yu M, Kim TH.
Presented at IEEE/CVF Computer Society Conference on Computer Vision and Pattern Recognition Workshops. 2023.