See clearly in the distance : representation learning GAN for low resolution object recognition

Abstract

Identifying tiny objects with extremely low resolution is generally considered a very challenging task even for human vision, due to limited information presented inside the object areas. There have been very limited attempts in recent years to deal with low-resolution recognition. The existing solutions rely on either generating super-resolution images or learning multi-scale features. However, their performance improvement becomes very limited, especially when the resolution becomes very low. In this paper, we propose a Representation Learning Generative Adversarial Network (<italic>RL</italic>-GAN) to generate super <italic>image representation</italic> that is optimized for recognition. Our solution deals with the classical vision task of object recognition in the distance. We evaluate our idea on the challenging task of low-resolution object recognition. Comparison of experimental results conducted on public and our newly created WIDER-SHIP datasets demonstrate the effectiveness of our <italic>RL</italic>-GAN, which improves the classification results significantly, with 10 & x2013;15 & x0025; gain on average, compared with benchmark solutions.