Vision Transformer (ViT) self-attention mechanism is characterized by feature
collapse in deeper layers, resulting in the vanishing of low-level visual
features. However, such features can be helpful to accurately represent and
identify elements within an image and increase the accuracy and robustness of
vision-based recognition systems. Following this rationale, we propose a novel
residual attention learning method for improving ViT-based architectures,
increasing their visual feature diversity and model robustness. In this way,
the proposed network can capture and preserve significant low-level features,
providing more details about the elements within the scene being analyzed. The
effectiveness and robustness of the presented method are evaluated on five
image classification benchmarks, including ImageNet1k, CIFAR10, CIFAR100,
Oxford Flowers-102, and Oxford-IIIT Pet, achieving improved performances.
Additionally, experiments on the COCO2017 dataset show that the devised
approach discovers and incorporates semantic and spatial relationships for
object detection and instance segmentation when implemented into spatial-aware
transformer models.

通过引入新的残差注意力学习方法来改善 Vision Transformer (ViT) 架构，包括增加视觉特征多样性和模型的鲁棒性，以提高识别系统的准确性。在五个图像分类基准测试中，通过对 ImageNet1k、CIFAR10、CIFAR100、Oxford Flowers-102 和 Oxford-IIIT Pet 数据集的评估，显示出所提方法的有效性和鲁棒性。此外，该方法在 COCO2017 数据集上的实验表明，在空间感知转换器模型中实现时，能发现和整合目标检测和目标实例分割的语义和空间关系。