About me

I am a Senior Data and Applied Scientist at Microsoft. Previously, I worked as an Applied Scientist at Amazon. I received my PhD from the Image Processing Group (GPI) at Universitat Politècnica de Catalunya. Over the course of my PhD, I interned at the National Institute of Informatics, the Massachusetts Institute of Technology, and Facebook AI Research.

[LinkedIn] [GScholar] [Github]

contact (at) amaiasalvador (dot) com


June 2022 - Present: Senior Data and Applied Scientist at Microsoft.

December 2019 - April 2022: Applied Scientist at Amazon.

Summer 2018: Research Internship. Facebook AI Research. Advisors: Adriana Romero and Michal Drozdzal.

Fall 2016: Research Internship. Massachusetts Institute of Technology. Advisor: Antonio Torralba.

Spring 2015: Research Internship. National Institute of Informatics. Advisor: Shin'ichi Satoh.

2015-2019: PhD in Computer Vision. Universitat Politècnica de Catalunya. Advisors: Xavier Giró-i-Nieto and Ferran Marqués.


June 2022: I joined the Core Search and AI team at Microsoft Barcelona.

March 2021: Our paper Revamping cross-modal recipe retrieval with hierarchical Transformers and self-supervised learning was accepted in CVPR 2021.

February 2020: I am one of the organizers of the Women in Computer Vision Workshop, to be held in conjunction with ECCV 2020 in Glasgow, UK.

December 2019: I joined the Computer Vision Team at Amazon Berlin.

July 2019: Our paper Recipe1M+: A Dataset for Learning Cross-Modal Embeddings for Cooking Recipes and Food Images was accepted in TPAMI 2019.

June 2019: I gave a talk at the 4th Summer School on Deep Learning for Computer Vision at UPC Barcelona. Slides here.

June 2019: I defended my Phd thesis, titled Computer Vision beyond the visible: Image understanding through language. Committee: Joost van der Weijer (CVC-UAB), Agata Lapedriza (UOC), Jasper Uijlings (Google AI), Laura Leal-Taixé (TUM), Javier Ruiz (UPC). Slides here.

June 2019: We received the best paper award at the CVPR 2019 DeepVision workshop for the paper Budget-aware Semi-Supervised Semantic and Instance Segmentation.

May 2019: Our paper on semi-supervised segmentation was accepted in CVPR-W DeepVision Workshop 2019.

February 2019: Two papers accepted in CVPR 2019 on video object segmentation and recipe generation.