Manuel Marques

I am a Researcher in Electrical and Computer Engineering Department at Instituto Superior Técnico and member of Signal and Image Processing Group at Institute for Robotics and Systems

manuel (at)

Instituto de Sistemas e Robótica
North Tower, 7th floor
Instituto Superior Técnico
Av. Rovisco Pais
1049-001 Lisboa - PORTUGAL
+351 218418083

profile photo

I'm interested in computer vision, machine learning, optimization, and image processing. In particular, my research is about inferring the shape of physical world from images. Please find below a selection of my publications or check my google scholar.

2D image head pose estimation via latent space regression under occlusion settings
José Celestino, Manuel Marques, Jacinto C. Nascimento, João Paulo Costeira,
Pattern Recognition, 2023

A novel deep learning approach for the problem of head pose estimation under occlusions based on latent space regression.

CYCLANDS: Cycling geo-located accidents, their details and severities
Miguel Costa, Manuel Marques, Carlos Roque, Filipe Moura,
Scientific Data, 2022
DATASET  |   code

CYCLANDS is a curated collection of 30 datasets on cycling crashes to lower the barrier in objective cycling research comprising nearly 1.6 M cycling accidents.

Rotation Averaging in a Split Second:
A Primal-Dual Method and a Closed-Form for Cycle Graphs

Gabriel Moreira, Manuel Marques, João Paulo Costeira,
IEEE International Conference on Computer Vision, 2021
ORAL presentation  |   poster  |   code

A very fast initialization-free primal-dual method and an optimal closed-form solution for rotation averaging in cycle graphs.

Recovery of Subspace Structure from High-Rank Data with Missing Entries
João Carvalho, Manuel Marques, João Paulo Costeira,
IEEE International Conference on Image Processing, 2019

A method to reconstruct and cluster incomplete high-dimensional data lying in a union of low-dimensional subspaces.

Motion analysis for people with cerebral palsy: A vision based approach
Mário Macedo, Alexandre Candeias, Manuel Marques,
IEEE International Conference on Rehabilitation Robotics, 2019

A methodology to classify motion of subjects with cerebral palsy based on RGB image sequences.

Vision Augmented Robot Feeding
Alexandre Candeias, Travers Rhodes, Manuel Marques, João Paulo Costeira, Manuela M. Veloso,
European Conference on Computer Vision - Workshop, 2018
FeedBot project  |   video

We evaluate how vision can be used to improve both food acquisition and delivery.

A Context Aware and Video-Based Risk Descriptor for Cyclists
Miguel Costa, Beatriz Quintino Ferreira, Manuel Marques,
IEEE International Conference on Intelligent Transportation Systems, 2017
project page

From smartphone sequences of images, we propose a new framework to assess the route risk from the cyclist perspective.

Subspace matching: Unique solution to point matching with geometric constraints
Manuel Marques, Marko Stosic, João Paulo Costeira,
IEEE International Conference on Computer Vision, 2009

We prove that the problem of matching a feature vector to a given subspace can be uniquely solved in a convex set.

Estimating 3D shape from degenerate sequences with missing data
Manuel Marques, João Paulo Costeira
Computer Vision and Image Understanding, February 2009

We propose to solve the structure from motion problem with degenerate data, introducing a new factorization algorithm that imposes the full scaled orthographic model in one single optimization procedure.