ReNet: A Recurrent Neural Network Based Alternative to Convolutional Networks
This work addresses object recognition for computer vision researchers by offering an incremental alternative to convolutional networks.
The authors tackled object recognition by proposing ReNet, a deep neural network architecture that replaces convolutional layers with recurrent neural networks sweeping across images, achieving results on MNIST, CIFAR-10, and SVHN that suggest it is a viable alternative to convolutional networks.
In this paper, we propose a deep neural network architecture for object recognition based on recurrent neural networks. The proposed network, called ReNet, replaces the ubiquitous convolution+pooling layer of the deep convolutional neural network with four recurrent neural networks that sweep horizontally and vertically in both directions across the image. We evaluate the proposed ReNet on three widely-used benchmark datasets; MNIST, CIFAR-10 and SVHN. The result suggests that ReNet is a viable alternative to the deep convolutional neural network, and that further investigation is needed.