Guillem Brasó

I am a second year PhD candidate working with Laura Leal-Taixé at the Dynamic Vision and Learning group, TU Munich.

Before starting my PhD, I obtained a Master's in Mathematics from TU Munich, while working on Multi-Object Tracking at my current lab. Before that, I got a Bachelor in Mathematics from the University of Barcelona. During my Bsc, I worked with Jordi Vitrià on explainable machine learning.

In my free time, I like to climb, play the piano, and cook.

Email  /  CV  /  Google Scholar  /  Twitter  /  Github

profile photo
News

[September 2022] Our journal extension of MPNTrack is accepted to IJCV!
[September 2022] Our paper on fully connected layers for low-data regimes is accepted to NeurIPS 2022!
[July 2022] PolarMOT, our new geometry-based 3D tracker, is accepted to ECCV 2022!
[May 2022] I received a CVPR 2022 outstanding reviewer award!
[March 2022] We are organizing the BMTT workshop at CVPR 2022. Check out our synth2real chalenges for tracking!
[March 2022] Just created this website!

Research

I have a general interest in Machine Learning and Computer Vision. Some of the tasks I focus on are detection, segmentation, tracking and human pose estimation. I am also broadly interested in leveraging ideas from classical graph-based approaches and optimization in combination with deep learning to solve vision problems.

blind-date Multi-Object Tracking and Segmentation via Neural Message Passing
Guillem Brasó*, Orçun Çetintas*, Laura Leal-Taixé
IJCV 2022

We build upon our neural solver for multi-object tracking to develop a joint model for tracking and segmentation with neural message passing. Our main contribution is an attention-based message passing framework that operates over segmentation masks and allows us to exploit synergies between data association and segmentation and SOTA results across multiple datasets.

paper | code
blind-date The Unreasonable Effectiveness of Fully-Connected Layers for Low-Data Regimes
Peter Kocsis, Peter Súkeník, Guillem Brasó, Matthias Niessner, Laura Leal-Taixé, Ismail Elezi
NeurIPS 2022

We show that augmenting modern classification architectures with a few fully connected layers during training significantly improves their generalization performance on low-data regimes. We propose a self-distillation mechanism that allows us to keep the number of parameter constant at test-time, while yielding universal improvements across backbones in datasets.

paper | website
blind-date PolarMot: How Far Can Geometric Relations Take Us in 3D Multi-Object Tracking?
Alexandr Kim, Guillem Brasó, Aljosa Osep, Laura Leal-Taixé
ECCV 2022

We propose an appearance-free 3D multi-object tracker that learns to associate 3D bounding boxes over time based on their relative geometric features and spatio-temporal relationships. Our model is light-weight and generalizes remarkably well among different locations and dataset.

paper | website | code
blind-date The Center of Attention: Center-Keypoint Attention for Multi-Person Pose Estimation
Guillem Brasó, Nikita Kister, Laura Leal-Taixé
ICCV 2021

We propose an multi-head attention-based framework for end-to-end bottom-up multi-person pose estimation. Our main idea is to detect both keypoints and object centers, and use cross-attention among them to group keypoints into human poses.

paper | video | code
blind-date MOTSynth: How Can Synthetic Data Help Pedestrian Detection and Tracking?
Matteo Fabbri, Guillem Brasó, Gianluca Maugeri, Orcun Cetintas, Riccardo Gasparini, Aljosa Osep, Simone Calderara, Laura Leal-Taixe, Rita Cucchiara
ICCV 2021

We introduce MOTSynth, a large-scale synthetic dataset for pedestrian tracking, detection, segmentation and 2D/3D keypoint estimation. We show that trackers and detectors trained on our dataset achieve impressive generalization results when tested on real data.

paper | dataset | video | code
blind-date Learning a Neural Solver for Multiple Object Tracking
Guillem Brasó, Laura Leal-Taixé
CVPR 2020   (Oral Presentation)

Inspired by classical graph-based multi-object tracking methods, we propose a neural message passing framework for data association in multi-object tracking. We present the first graph-structured deep learning module for tracking, and achieve SOTA results on the MOTChallenge benchmarks.

paper | video | code

This website's source code is from Jon Barron.