模仿视觉皮层机制设计神经网络滤波器，提升被遮挡图像的边缘补全能力。
Bioinspired CNNs for border completion in occluded images
- 借鉴视觉皮层数学模型设计CNN滤波器
- 在三种数据集上对条纹和网格遮挡均有性能提升
- 适用于图像修复与鲁棒性增强场景
我们利用视觉皮层中边缘补全问题的数学建模，设计了能够增强图像遮挡鲁棒性的卷积神经网络（CNN）滤波器。在三种遮挡数据集（MNIST、Fashion-MNIST 和 EMNIST）上，针对条纹和网格两种遮挡类型评估了我们的CNN架构BorderNet。在所有情况下，BorderNet均表现出更优性能，增益程度取决于遮挡严重程度和数据集类型。
原文摘要 · Abstract (English)
We exploit the mathematical modeling of the border completion problem in the visual cortex to design convolutional neural network (CNN) filters that enhance robustness to image occlusions. We evaluate our CNN architecture, BorderNet, on three occluded datasets (MNIST, Fashion-MNIST, and EMNIST) under two types of occlusions: stripes and grids. In all cases, BorderNet demonstrates improved performance, with gains varying depending on the severity of the occlusions and the dataset.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。