Marco De Nadai

Marco De Nadai

Research Scientist


I am an Applied Scientist at Zalando, Berlin, Germany. In Zalando, I focus on outfit generation to inspire people and suggest new fashion styles. I also work on computer vision, unsupervised learning, and human behavioural understanding through multi-modal data. My work includes modelling human mobility and perception, photorealistic object manipulation and animation, video synthesis and image-to-image translation. My work has been published in multiple top conference papers and journals.

Previously, I was a Research Scientist at the Bruno Kessler Foundation (FBK), Trento, Italy. I am a PhD graduate from the University of Trento, Italy, where I worked with Bruno Lepri, at the Mobs Lab, and Nicu Sebe at the Multimedia and Human Understanding Group MHUG.

I collaborated with numerous international institutions including MIT, MIT Media Lab, Nokia Bell Labs, Vodafone Group and Data-pop Alliance. Before joining FBK, I was a research intern at Vodafone, London, UK, and worked with Angelo Cardoso and Nuria Oliver. I also visited the Massachusetts Institute of Technology (MIT) and worked with Marta C. Gonzalez and César Hidalgo.

Download my resumé.

  • Human Behavioural Understanding
  • Computer Vision
  • Generative Networks
  • Artificial Intelligence
  • PhD in Computer Science (cum laude), 2019

    University of Trento

  • MSc in Computer Science (cum laude), 2015

    University of Trento

  • BSc in Computer Science, 2012

    University of Udine

Recent Publications

Quickly discover relevant content by filtering publications.
(2021). Efficient Training of Visual Transformers with Small-Size Datasets. NeurIPS ‘21.


(2021). Click to Move: Controlling Video Generation with Sparse Motion. ICCV 2021.

PDF Code

(2021). Smoothing the Latent Style Space for Unsupervised Image-to-Image Translation. CVPR 2021.

PDF Cite Code

(2021). Living in a pandemic: adaptation of individual mobility and social activity in the US. Preprint.


(2021). The agglomeration and dispersion dichotomy of human settlements on Earth. Preprint.