Guillermo Garcia-Hernando Research Scientist at Niantic, Inc.

About

I'm a researcher at Niantic's AR Research team led by Gabriel Brostow and based in London, UK. Previously, I spent a couple of years as a postdoc at Imperial College London in collaboration with Samsung Research. I completed my Ph.D. in computer vision and machine learning in 2017 supervised by Tae-Kyun Kim also at Imperial. In a previous life, I was in Paris (MSc./Diplôme d'Ingénieur at Télécom Paris) and in Barcelona (BSc. at Technical University of Catalonia - UPC).

I particularly enjoy teaching, supervising students and disseminating research to audiences of different backgrounds and ages. I've been involved in the organization of academic events such as BMVC 2017, different workshops at CVPR, ICCV and ECCV conferences and also had the chance of teaching at the Nepal Winter School in AI 2018.

Selected publications

  • Predicting Visual Overlap of Images Through Interpretable Non-Metric Box Embeddings

    Anita Rau, Guillermo Garcia-Hernando, Danail Stoyanov, Gabriel J. Brostow and Daniyar Turmukhambetov.

    ECCV 2020 (spotlight)

    Paper Project
  • Measuring Generalisation to Unseen Viewpoints, Articulations, Shapes and Objects for 3D Hand Pose Estimation under Hand-Object Interaction

    Anil Armagan, Guillermo Garcia-Hernando, Seungryul Baek, Vincent Lepetit, Tae-Kyun Kim et al.

    ECCV 2020

    Paper Project Dataset
  • Introducing Pose Consistency and Warp-Alignment for Self-Supervised 6D Object Pose Estimation in Color Images

    Juil Sock, Guillermo Garcia-Hernando, Anil Armagan and Tae-Kyun Kim

    3DV 2020 (oral)

    Paper
  • Physics-Based Dexterous Manipulations with Estimated Hand Poses and Residual Reinforcement Learning

    Guillermo Garcia-Hernando, Edward Johns and Tae-Kyun Kim.

    IROS 2020

    Paper Project
  • Active 6D Multi-Object Pose Estimation in Cluttered Scenarios with Deep Reinforcement Learning

    Juil Sock, Guillermo Garcia-Hernando and Tae-Kyun Kim.

    IROS 2020

    Paper
  • First-Person Hand Action Benchmark with RGB-D Videos and 3D Hand Pose Annotations

    Guillermo Garcia-Hernando, Shanxin Yuan, Seungryul Baek and Tae-Kyun Kim.

    CVPR 2018

    Paper Project Dataset
  • Depth-Based 3D Hand Pose Estimation: From Current Achievements to Future Goals

    Shanxin Yuan, Guillermo Garcia-Hernando, Bjorn Stenger, Tae-Kyun Kim et al.

    CVPR 2018 (spotlight)

    Paper Project
  • Task-Oriented Hand Motion Retargeting for Dexterous Manipulation Imitation

    Dafni Antotsiou, Guillermo Garcia-Hernando and Tae-Kyun Kim.

    ECCV 2018 (workshop)

    Paper Project
  • Transition Forests: Learning Discriminative Temporal Transitions for Action Recognition and Detection

    Guillermo Garcia-Hernando and Tae-Kyun Kim.

    CVPR 2017

    Paper
  • Transition Hough Forest for Trajectory-based Action Recognition

    Guillermo Garcia-Hernando, Hyung Jin Chang, Ismael Serrano, Oscar Deniz and Tae-Kyun Kim.

    WACV 2016

    Paper Dataset
  • Spatio-Temporal Hough Forest for Efficient Detection-Localisation-Recognition of Fingerwriting in Egocentric Camera

    Hyung Jin Chang*, Guillermo Garcia-Hernando*, Danhang Tang and Tae-Kyun Kim.

    CVIU 2016

    Paper Dataset
  • Understanding egocentric human actions with temporal decision forests

    Guillermo Garcia-Hernando

    Imperial College London, Ph.D. Thesis. 2017

    Paper