A Discriminative Channel Diversification Network for Image Classification
This work addresses efficiency and accuracy in image classification for computer vision applications, but it is incremental as it builds on existing channel attention methods with a novel module placement.
The paper tackles the problem of high model complexity and computation cost in channel attention mechanisms for image classification by proposing a lightweight channel diversification block that enhances global context and focuses on discriminative features. The result is an average performance improvement of 3% on datasets like CIFAR-10, SVHN, and Tiny-ImageNet.
Channel attention mechanisms in convolutional neural networks have been proven to be effective in various computer vision tasks. However, the performance improvement comes with additional model complexity and computation cost. In this paper, we propose a light-weight and effective attention module, called channel diversification block, to enhance the global context by establishing the channel relationship at the global level. Unlike other channel attention mechanisms, the proposed module focuses on the most discriminative features by giving more attention to the spatially distinguishable channels while taking account of the channel activation. Different from other attention models that plugin the module in between several intermediate layers, the proposed module is embedded at the end of the backbone networks, making it easy to implement. Extensive experiments on CIFAR-10, SVHN, and Tiny-ImageNet datasets demonstrate that the proposed module improves the performance of the baseline networks by a margin of 3% on average.