Loading...

Research collaboration between Qualcomm and UvA

Our Mission

UvA
Qualcomm

The mission of the QUVA-lab is to perform world-class research on deep vision. Such vision strives to automatically interpret with the aid of deep learning what happens where, when and why in images and video. Deep learning is a form of machine learning with neural networks, loosely inspired by how neurons process information in the brain. Research projects in the lab focus on learning to recognize objects in images from a single example, personalized event detection and summarization in video, and privacy preserving deep learning. The research is published in the best academic venues and secured in patents.

Learn More

Highlights

QUVA Colloqium: Dima Damen

Apr 13, 2022

On Apr 13, we hosted Dima Damen giving a talk about her work on "Video Understanding - An Egocentric Perspective" in our Qualcomm-UvA Deep Vision Seminars.

Watch the recording

QUVA Colloqium: Andreas Geiger

Mar 17, 2022

On Mar 17, we hosted Andreas Geiger giving a talk about his work on "Neural Implicit Representations for 3D Vision and Beyond" in our Qualcomm-UvA Deep Vision Seminars.

Watch the recording

QUVA Colloqium: Olga Russakovsky

Jan 27, 2022

On Jan 27, we hosted Olga Russakovsky giving a talk about her work on "Fairness in Visual Recognition" in our Qualcomm-UvA Deep Vision Seminars.

Watch the recording

QUVA Colloqium

The Qualcomm-UvA Deep Vision Seminars is an Amsterdam meetup for people that are passionate about AI, machine learning, deep learning and computer vision. Our guest speakers come both from industry and academia, working for organizations such as DeepMind, Oxford University, Toyota Research Center and more. Subscribe to our Meetup page to be notified about upcoming talks, or watch past presentations here.

Work with the QUVA Lab

Interested in working with the QUVA Lab? We have open positions! See all here.

PhD Position | Geometric Deep Learning of Space and Time

Efstratios Gavves, Taco Cohen

In this position we will study the theory and applications of geometry in various aspects of deep neural networks, focusing on spatial, as well as on temporal data and dynamics.

Details and application

Master Thesis Intern | Does 3D modeling help image representation learning?

Pengwan Yang, Yuki Asano

Our goal is to find out whether 3D modelling approaches such as NERF can be used for the task of self-supervised representation learning.

Details and application

Master Thesis Intern | Self-Supervised Learning on Point Clouds by Free Form Deformation

Pengwan Yang, Yuki Asano

The goal of this project is to apply state of the art self-supervised learning methods on pointclouds.

Details and application

Master Thesis Intern | Dense self-supervised test-time training for recycling large-scale pretrained visual encoders

Mohammadreza Salehi, Yuki Asano

The goal of this project is to adapt self-supervised pretrained models for a dense prediction task using only a single training sample and no labels.

Details and application

Master Thesis Intern | Spatial prompt tuning for temporal action recognition

Mohammadreza Salehi, Yuki Asano

The goal of this project is to design a prompt tuning method to employ CLIP’s knowledge to improve the performance on action recognition tasks.

Details and application

Master Thesis Intern | Differential Privacy guarantees in Deep Learning

Rob Romijnders

This project focuses on the privacy aspects of training neural networks.

Details and application

Master Thesis Intern | Sharing deep learning models with privacy guarantees

Rob Romijnders

This project focuses on the privacy aspects of sharing neural networks trained on sensitive data.

Details and application

Master Thesis Intern | Contact tracing in a COVID-like pandemic: about privacy

Rob Romijnders

In this project, we address the problem of privacy in such contact tracing: one wants to prevent at all costs that COVID scores of individuals become freely available.

Details and application

Master Thesis Intern | Adaptive sampling for continuous Group Equivariant CNNs

Gabriele Cesa

The focus of this project is building deep learning models equivariant to continuous transformations by relying on an adaptive sampling strategy.

Details and application

Master Thesis Intern | Learning the Degree of Equivariance in Steerable CNNs

Gabriele Cesa

The focus of this project is learning the set of transformations a deep learning model should be equivariant to.

Details and application

Master Thesis Intern | Analyzing Self-Supervised Methods for Video Representation Learning

Michael Dorkenwald, Fida Mohammad

The goal of this project is to find out what is actually learned with self-supervised methods for video representation learning.

Details and application

Contact

University of Amsterdam

  • Science Park 900 (LAB42, Rooms L4.35/36)
    1098XH Amsterdam
    The Netherlands

Secretary

  • Virgine Mes 
      v.m.j.mes at uva dot nl
      0205256409

  •  Twitter: @quvalab