D-PCN: Parallel Convolutional Networks for Image Recognition via a Discriminator
This work addresses the need for better feature extraction in image recognition, particularly for tasks like fine-grained classification and segmentation, but it appears incremental as it builds on existing CNN architectures.
The paper tackles the problem of enhancing feature extraction in CNNs for image recognition by introducing D-PCN, a framework with parallel CNNs and a discriminator that drives complementary representations, resulting in state-of-the-art performance on CIFAR-100 and improvements on other datasets.
In this paper, we introduce a simple but quite effective recognition framework dubbed D-PCN, aiming at enhancing feature extracting ability of CNN. The framework consists of two parallel CNNs, a discriminator and an extra classifier which takes integrated features from parallel networks and gives final prediction. The discriminator is core which drives parallel networks to focus on different regions and learn complementary representations. The corresponding joint training strategy is introduced which ensures the utilization of discriminator. We validate D-PCN with several CNN models on two benchmark datasets: CIFAR-100 and ImageNet32x32, D-PCN enhances all models. In particular it yields state of the art performance on CIFAR-100 compared with related works. We also conduct visualization experiment on fine-grained Stanford Dogs dataset and verify our motivation. Additionally, we apply D-PCN for segmentation on PASCAL VOC 2012 and also find promotion.