Search Results for author: Martin Martens

Found 1 papers, 1 papers with code

Multi-Stage Reinforcement Learning For Object Detection

1 code implementation15 Oct 2018 Jonas Koenig, Simon Malberg, Martin Martens, Sebastian Niehaus, Artus Krohn-Grimberghe, Arunselvan Ramaswamy

We compare an approach that is based purely on zoom actions with one that is extended by a second refinement stage to fine-tune the bounding box after each zoom step.

Object object-detection +3

Cannot find the paper you are looking for? You can Submit a new open access paper.