We consider how image super resolution (SR) can contribute to an object detection task in low-resolution images. Intuitively, SR gives a positive impact on the object detection task. While several previous works demonstrated that this intuition is correct, SR and detector are optimized independently in these works. This paper proposes a novel framework to train a deep neural network where the SR sub-network explicitly incorporates a detection loss in its training objective, via a tradeoff with a traditional detection loss. This end-to-end training procedure allows us to train SR preprocessing for any differentiable detector. We demonstrate that our task-driven SR consistently and significantly improves accuracy of an object detector on low-resolution images for a variety of conditions and scaling factors.
@article{arxiv.1803.11316,
title = {Task-Driven Super Resolution: Object Detection in Low-resolution Images},
author = {Muhammad Haris and Greg Shakhnarovich and Norimichi Ukita},
journal= {arXiv preprint arXiv:1803.11316},
year = {2018}
}