Higher levels of machine intelligence demand alignment with human perception and cognition. Deep neural networks (DNN) dominated machine intelligence have demonstrated exceptional performance across various real-world tasks. Nevertheless, recent evidence suggests that DNNs fail to perceive illusory contours like the abutting grating, a discrepancy that misaligns with human perception patterns. Departing from previous works, we propose a novel deep network called illusory contour perception network (ICPNet) inspired by the circuits of the visual cortex. In ICPNet, a multi-scale feature projection (MFP) module is designed to extract multi-scale representations. To boost the interaction between feedforward and feedback features, a feature interaction attention module (FIAM) is introduced. Moreover, drawing inspiration from the shape bias observed in human perception, an edge detection task conducted via the edge fusion module (EFM) injects shape constraints that guide the network to concentrate on the foreground. We assess our method on the existing AG-MNIST test set and the AG-Fashion-MNIST test sets constructed by this work. Comprehensive experimental results reveal that ICPNet is significantly more sensitive to abutting grating illusory contours than state-of-the-art models, with notable improvements in top-1 accuracy across various subsets. This work is expected to make a step towards human-level intelligence for DNN-based models.

翻译：更高层次的机器智能要求与人类感知和认知对齐。以深度神经网络（DNN）为主导的机器智能已在多种现实任务中展现出卓越性能。然而，最新证据表明，DNN无法像人类一样感知对接光栅等错觉轮廓，这一差异导致其与人类感知模式不一致。不同于以往研究，我们提出一种受视觉皮层回路启发的全新深度网络——错觉轮廓感知网络（ICPNet）。在ICPNet中，设计了一个多尺度特征投影（MFP）模块以提取多尺度表征。为增强前馈与反馈特征之间的交互，引入了一种特征交互注意力模块（FIAM）。此外，受人类感知中形状偏置的启发，通过边缘融合模块（EFM）执行边缘检测任务，注入形状约束以引导网络聚焦于前景。我们在现有AG-MNIST测试集及本研究构建的AG-Fashion-MNIST测试集上评估了所提方法。综合实验结果表明，ICPNet对对接光栅错觉轮廓的敏感性显著优于现有最优模型，并在多个子集上取得了Top-1精度的显著提升。本工作有望推动基于DNN的模型向人类级智能迈进。