Bio-Vision-Inspired Spiking Neural Networks for Object Detection with Event Cameras

Retina-like event cameras and brain-inspired Spiking Neural Networks (SNNs) demonstrate exceptional energy efficiency through bio-inspired sensing and computation. While SNNs are naturally well-suited to the asynchronous nature of event data, their practical applications face the following challenges: sensitivity to noise, dense representations that disrupt spike pathways, and insufficient multi-scale feature perception. To address the aforementioned challenges, we propose a bio-vision-inspired object detection method motivated by biological (bio) vision systems. First, at the micro level, this paper proposes a noise-filtering STATNF-Neuron architecture to address the current sensitivity of basic neurons to noise. Based on STATNF-Neurons, the paper introduces two bio-vision-inspired macro-structures: Events-to-Spikes Representation (E2S), which preserves spiking characteristics while mimicking the memory and noise-filtering abilities of retinal neurons; Bidirectional Multi-Scale Spiking Network (BiSNet), which simulates cortical information flow pathways to integrate multi-scale features in both directions, enhancing the network's ability to perceive information at multiple scales. Extensive experiments show that the proposed bio-vision-inspired method achieving state-of-the-art performance. Notably, it reaches 96.1% accuracy on NCAR, 63.5% mAP(50) on N-Caltech101, and 69.1% mAP(50) on Gen1.

Traditional cameras and artificial intelligence often struggle to detect objects under fast-moving or extreme lighting conditions, and they typically consume a significant amount of power. Inspired by the mechanism of the human eye, event cameras offer a novel solution: they capture only the dynamic changes in a scene, thereby saving energy and drastically increasing processing speed. However, processing this unique data with brain-inspired artificial intelligence (known as Spiking Neural Networks) often suffers from background noise interference, which not only reduces detection accuracy but also causes energy waste. To address this issue, we designed a brand new system that closely mimics how the biological retina and visual cortex naturally filter out noise and process information at multiple scales. Our brain-inspired approach significantly improves object detection accuracy in highly challenging scenarios while keeping the system's energy consumption at an extremely low level.