Visual recognition research often assumes a sufficient resolution of the region of interest (ROI). That is usually violated in practical scenarios, inspiring us to explore the general Very Low Resolution Recognition (VLRR) problem. Typically, the ROI in a VLRR problem can be smaller than $16 \times 16$ pixels, and is challenging to be recognized even by human experts. We attempt to solve the VLRR problem using deep learning methods. Taking advantage of techniques primarily in super resolution, domain adaptation and robust regression, we formulate a dedicated deep learning method and demonstrate how these techniques are incorporated step by step. That leads to a series of well-motivated and powerful models. Any extra complexity due to the introduction of a new model is fully justified by both analysis and simulation results. The resulting \textit{Robust Partially Coupled Networks} achieves feature enhancement and recognition simultaneously, while allowing for both the flexibility to combat the LR-HR domain mismatch and the robustness to outliers. Finally, the effectiveness of the proposed models is evaluated on three different VLRR tasks, including face identification, digit recognition and font recognition, all of which obtain very impressive performances.
from cs.AI updates on arXiv.org http://ift.tt/1ZIxC6K
via IFTTT
No comments:
Post a Comment