TY  - CONF
AU  - Kamal Nasrollahi
AU  - Sergio Escalera
AU  - P. Rasti
AU  - Gholamreza Anbarjafari
AU  - Xavier Baro
AU  - Hugo Jair Escalante
AU  - Thomas B. Moeslund
A2  - IPTA
PY  - 2015//
TI  - Deep Learning based Super-Resolution for Improved Action Recognition
BT  - 5th International Conference on Image Processing Theory, Tools and Applications IPTA2015
SP  - 67
EP  - 72
N2  - Action recognition systems mostly work with videos of proper quality and resolution. Even most challenging benchmark databases for action recognition, hardly include videos of low-resolution from, e.g., surveillance cameras. In videos recorded by such cameras, due to the distance between people and cameras, people are pictured very small and hence challenge action recognition algorithms. Simple upsampling methods, like bicubic interpolation, cannot retrieve all the detailed information that can help the recognition. To deal with this problem, in this paper we combine results of bicubic interpolation with results of a state-ofthe-art deep learning-based super-resolution algorithm, through an alpha-blending approach. The experimental results obtained on down-sampled version of a large subset of Hoolywood2 benchmark database show the importance of the proposed system in increasing the recognition rate of a state-of-the-art action recognition system for handling low-resolution videos.
L1  - http://refbase.cvc.uab.es/files/NER2015.pdf
UR  - http://dx.doi.org/10.1109/IPTA.2015.7367098
N1  - HuPBA;MV
ID  - Kamal Nasrollahi2015
ER  -