Erickson R. Nascimento

bio

teaching

research

publications

collaborators

Erickson R. Nascimento


Department of Computer Science at Federal University of Minas Gerais (UFMG), Brazil
Member of the Computer Vision and Robotics (VeRLab) research group
Supported by a CNPq research fellowship
Google Scholar, CV Lattes (in Portuguese)


[NEW] Paper accepted at IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) 2024 - XFeat: Accelerated Features for Lightweight Image Matching

[NEW] Paper accepted at IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) 2023 - Enhancing Deformable Local Features by Jointly Learning to Detect and Describe Keypoints

[NEW] Paper accepted at IEEE Transactions on Pattern Analysis and Machine Intellignece (TPAMI) 2023 - Text-driven video acceleration: A weakly-supervised reinforment learning method [arXiv, DOI]

[NEW] Paper accepted at British Machine Vision Conference (BMVC) 2022 - Semantic Segmentation under Adverse Conditions: A Weather and Nighttime-aware Synthetic Data-based [arXiv]

[NEW] Paper accepted at Computer Vision and Image Understanding (CVIU) 2022 - Learning Geodesic-Aware Local Features from RGB-D Images [arXiv, DOI]

[NEW] Paper accepted at WACV 2022 - Creating and reenacting controllable 3D Humans with differentiable rendering

[NEW] Paper accepted at NeurIPS 2021 - Extracting Deformation-Aware Local Features by Learning to Deform - Visit the project page for code and data!

[NEW] Paper accepted to the International Journal of Computer Vision (IJCV) - A Shape-Aware Retargeting Approach to Transfer Human Motion and Appearance in Monocular Videos - Visit the project page for more information!

[NEW] Oustanding reviewer of CVPR 2021

[NEW] Paper accepted to ICRA 2021! - Anytime Fault-tolerant Adaptive Routing for Multi-Robot Teams

[NEW] Paper accepted to IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI) - A Sparse Sampling-based framework for Semantic Fast-Forward of First-Person Videos - Visit the project page for more information!

Paper accepted to Computer & Graphics - Learning to dance: A graph convolutional adversarial network to generate realistic dance motions from audio - Visit the project page for more information!

Paper accepted to CVPR 2020 - Straight to the Point: Fast-forwarding Videos via Reinforcement Learning Using Textual Data - Visit the project page for more information!

Paper accepted to CVPR 2020 EPIC Workshop - A gaze driven fast-forward method for first-person videos - Visit the project page for more information!

Two papers accepted to WACV 2020:


Paper accepted to ICCV 2019 - GEOBIT: A Geodesic-Based Binary Descriptor Invariant to Non-Rigid Deformations for RGB-D images: Visit the project page for more information!

Paper accepted to WACV 2019 - Prototypicality effects in global semantic description of objects: Visit the project page for more information!

Paper accepted to CVPR 2018 - A Weighted Sparse Sampling and Smoothing Frame Transition Approach for Semantic Fast-Forward First-Person Videos: Visit the project page for more information!

Paper accepted to CVPR 2018 Sight and Sound Workshop - Fast forwarding Egocentric Videos by Listening and Watching: Visit the project page for more information!

Guest Editor of the Special Issue on Feature representations for medical images and activity understanding - Journal of Visual Communication and Image Representationa (JVCI)

Publication chair of ICAR 2019, Program Co-chair of SIBGRAPI 2017


Member of the program committee for:


Research areas

  • Egocentric videos
  • Multimodal learning
  • Nonrigid descriptors
  • Participating Media
  • 3D Reconstruction and Modeling
  • Image and Geometrical Descriptors

How to find me

  • erickson [at] dcc [dot] ufmg [dot] br
  • Room 6318, DCC - ICEx - UFMG
  • Phone (+55 31) 3409-7597
VeRLab Department of Computer Science Federal University of Minas Gerais