Predictive Modeling of Equine Activity Budgets Using a 3D Skeleton Reconstructed from Surveillance Recordings

no code implementations • 8 Jun 2023 • Ernest Pokropek, Sofia Broomé, Pia Haubro Andersen, Hedvig Kjellström

In this work, we present a pipeline to reconstruct the 3D pose of a horse from 4 simultaneous surveillance camera recordings.

Pose Estimation

Recur, Attend or Convolve? On Whether Temporal Modeling Matters for Cross-Domain Robustness in Action Recognition

1 code implementation • 22 Dec 2021 • Sofia Broomé, Ernest Pokropek, Boyu Li, Hedvig Kjellström

Most action recognition models today are highly parameterized, and evaluated on datasets with appearance-wise distinct classes.

Action Recognition Domain Generalization +1

The World as a Graph: Improving El Niño Forecasts with Graph Neural Networks

2 code implementations • 11 Apr 2021 • Salva Rühling Cachay, Emma Erickson, Arthur Fender C. Bucker, Ernest Pokropek, Willa Potosnak, Suyash Bire, Salomey Osei, Björn Lütjens

In comparison, graph neural networks (GNNs) are capable of modeling large-scale spatial dependencies and are more interpretable due to the explicit modeling of information flow through edge connections.

Graph Neural Networks for Improved El Niño Forecasting

1 code implementation • 2 Dec 2020 • Salva Rühling Cachay, Emma Erickson, Arthur Fender C. Bucker, Ernest Pokropek, Willa Potosnak, Salomey Osei, Björn Lütjens

Deep learning-based models have recently outperformed state-of-the-art seasonal forecasting models, such as for predicting El Ni\~no-Southern Oscillation (ENSO).

Multivariate Time Series Forecasting Spatio-Temporal Forecasting