Locally Scale-Invariant Convolutional Neural Networks. Kanazawa, A., Sharma, A., & Jacobs, D. ArXiv e-prints, December, 2014.
Locally Scale-Invariant Convolutional Neural Networks [link]Paper  bibtex   
@article{Kanazawa:2014ub,
author = {Kanazawa, Angjoo and Sharma, Abhishek and Jacobs, David},
title = {{Locally Scale-Invariant Convolutional Neural Networks}},
journal = {ArXiv e-prints},
year = {2014},
volume = {cs.CV},
month = dec,
annote = {a very naive approach to mutliscale.

See Figure 1 and 2.

I just suspect the usefulness of this construct on standard datasets... They don't even have experiments on MNIST....},
keywords = {deep learning},
read = {Yes},
rating = {2},
date-added = {2017-05-05T17:11:21GMT},
date-modified = {2017-05-05T17:14:55GMT},
url = {http://arxiv.org/abs/1412.5104},
local-url = {file://localhost/Users/yimengzh/Documents/Papers3_revised/Library.papers3/Articles/2014/Kanazawa/arXiv%202014%20Kanazawa.pdf},
file = {{arXiv 2014 Kanazawa.pdf:/Users/yimengzh/Documents/Papers3_revised/Library.papers3/Articles/2014/Kanazawa/arXiv 2014 Kanazawa.pdf:application/pdf}},
uri = {\url{papers3://publication/uuid/2C26B012-F088-413D-86F9-9EC429DF45FF}}
}

Downloads: 0