Advancing event-driven vision through spiking neural networks (SNNs) is crucial to empowering high-speed and efficient perception. While directly converting the pre-trained artificial neural networks (ANNs) - by replacing the non-linear activation with spiking neurons - can provide SNNs with good performance, the resultant SNNs typically demand long timesteps and high energy consumption to achieve their optimal performance. To address this challenge, we introduce the burst-spike mechanism inspired by the biological nervous system, allowing multiple spikes per timestep to reduce conversion errors and produce low-latency SNNs. To further bolster this enhancement, we leverage the Pareto Frontier-driven algorithm to reallocate burst-firing patterns. Moreover, to reduce energy consumption during the conversion process, we propose a sensitivity-driven spike compression technique, which automatically locates the optimal threshold ratio according to layer-specific sensitivity. Extensive experiments demonstrate our approach outperforms state-of-the-art SNN methods, showcasing superior performance and reduced energy usage across classification and object detection. Our code will be available at https://github.com/bic-L/burst-ann2snn.

翻译：推进通过脉冲神经网络（SNNs）实现事件驱动视觉对于赋能高速高效感知至关重要。虽然直接转换预训练的人工神经网络（ANNs）——用脉冲神经元替代非线性激活函数——能够使SNNs获得良好性能，但此类SNNs通常需要较长的时间步长和高能耗才能达到最优性能。为应对这一挑战，我们受生物神经系统启发引入爆发式脉冲机制，允许每个时间步产生多个脉冲以减少转换误差并构建低延迟SNNs。为进一步强化这一改进，我们利用帕累托前沿驱动算法重新分配爆发发放模式。此外，为降低转换过程中的能耗，我们提出一种灵敏度驱动的脉冲压缩技术，该技术根据层特定灵敏度自动定位最优阈值比率。大量实验表明，我们的方法在分类和目标检测任务上均优于现有最先进的SNN方法，展现出卓越性能与更低能耗。我们的代码将在 https://github.com/bic-L/burst-ann2snn 开源。