.png){kind=icon}
.png){kind=icon}
Return
Circulant Binary Convolutional Networks for Object Recognition
.png){kind=icon}
PRE
AI
DOI：10.1109/JSTSP.2020.2969516.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
The rapidly decreasing computation and memory cost has recently driven the success of many applications in the field of deep learning. Practical applications of deep learning in resource-limited hardware, such as embedded devices and smart phones, however, remain challenging. For binary convolutional networks, the reason lies in the degraded representation caused by binarizing full-precision filters. To address this problem, we propose new circulant filters (CiFs) and a circulant binary convolution (CBConv) to enhance the capacity of binarized convolutional features via our circulant back propagation (CBP). The CiFs can be easily incorporated into existing deep convolutional neural networks (DCNNs), which leads to new Circulant Binary Convolutional Networks (CBCNs). Extensive experiments confirm that the performance gap between the 1-bit and full-precision DCNNs is minimized by increasing the filter diversity, which further increases the representational ability in our networks. Our experiments on ImageNet show that CBCNs achieve 61.4% top-1 accuracy with ResNet18. Compared to the state-of-the-art such as XNOR, CBCNs can achieve up to 10% higher top-1 accuracy with more powerful representational ability. Also, CBCNs approximately achieve a storage reduction about 32 times. In particular, our method shows strong generalization on the object recognition task, i.e., face recognition, facial expression recognition and person re-identification.
Keywords:
Convolution
Backpropagation
Indexes
Face recognition
Task analysis
Computational modeling
Training
Circulant binary convolutional networks
1-bit DCNNs
circulant back propagation
circulant filters
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
IF:
13.7
Papers:
1.9K
Citations:
1.1W
Organization
View more organizations