Kernel Pooling for Convolutional Neural Networks
Research output: Contribution to journal › Conference article › Research › peer-review
Convolutional Neural Networks (CNNs) with Bilinear Pooling, initially in their full form and later using compact representations, have yielded impressive performance gains on a wide range of visual tasks, including fine-grained visual categorization, visual question answering, face recognition, and description of texture and style. The key to their success lies in the spatially invariant modeling of pairwise (2nd order) feature interactions. In this work, we propose a general pooling framework that captures higher order interactions of features in the form of kernels. We demonstrate how to approximate kernels such as Gaussian RBF up to a given order using compact explicit feature maps in a parameter-free manner. Combined with CNNs, the composition of the kernel can be learned from data in an endto- end fashion via error back-propagation. The proposed kernel pooling scheme is evaluated in terms of both kernel approximation error and visual recognition accuracy. Experimental evaluations demonstrate state-of-the-art performance on commonly used fine-grained recognition datasets.
Original language | English |
---|---|
Journal | Proceedings - 30th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2017 |
Pages (from-to) | 3049-3058 |
Number of pages | 10 |
DOIs | |
Publication status | Published - 2017 |
Externally published | Yes |
Event | 30th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2017 - Honolulu, United States Duration: 21 Jul 2017 → 26 Jul 2017 |
Conference
Conference | 30th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2017 |
---|---|
Country | United States |
City | Honolulu |
Period | 21/07/2017 → 26/07/2017 |
Bibliographical note
Funding Information:
This work was supported in part by Google Focused Re-searchAward,AWS CloudCreditsforResearch, Microsoft Research Award and a Facebook equipment donation.
Publisher Copyright:
©2017 IEEE.
ID: 301826599