Statistically-Motivated Second-order Pooling

Second-order pooling, a.k.a. bilinear pooling, has proven effective for deep learning based visual recognition. However, the resulting second-order networks yield a final representation that is orders of magnitude larger than that of standard, first-order ones, making them memory-intensive and cumbersome to deploy. Here, we introduce a general, parametric compression strategy that can produce more compact representations than existing compression techniques, yet outperform both compressed and uncompressed second-order models. Our approach is motivated by a statistical analysis of the network’s activations, relying on operations that lead to a Gaussian-distributed final representation, as inherently used by first-order deep networks. As evidenced by our experiments, this lets us outperform the state-of-the-art first-order and second-order models on several benchmark recognition datasets.


Presented at:
European Conference on Computer Vision 2018 (ECCV2018), Munich, Germany, September 8-14, 2018
Year:
2018
Keywords:
Laboratories:




 Record created 2018-08-22, last modified 2018-09-13

PREPRINT:
Download fulltext
PDF

Rate this document:

Rate this document:
1
2
3
 
(Not yet reviewed)