发表机构
Indian Institute of Technology Indore; Thadomal Shahani Engineering College, University of Mumbai; Indian Institute of Technology Kharagpur; Vellore Institute of Technology Chennai; Indiana University Indianapolis（印度理工学院印多尔分校; 孟买大学塔多马尔·沙哈尼工程学院; 印度理工学院克勒格布尔分校; 韦洛尔理工学院金奈分校; 印第安纳大学印第安纳波利斯分校）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
提出主动脉冲感知（ASP），以LIF膜电位为置信状态实现3D点云识别的迭代决策，在ModelNet等数据集取得结果，机制可迁移至密集预测，能耗可降2.8倍至1.35倍。
AI 中文摘要
脉冲点云网络通常以固定的、与输入无关的顺序扫描空间，这使得脉冲计算最独特的资源——膜电位的时间演化——未被用作决策的位点。主动脉冲感知（Active Spiking Perception, ASP）将3D识别重新表述为迭代决策过程，其中网络自身的泄漏整合发放（leaky integrate-and-fire, LIF）膜电位（被解读为对类别的运行置信度）选择下一个要观察的块并触发基于置信度边界的提前退出。轻量级切片选择策略从膜状态和预计算的几何描述符中对未访问的最远点采样块进行评分，通过直通Gumbel-Softmax进行端到端训练，推理时简化为argmax，仅增加约2%的骨干网络参数。我们证明，泄漏积分是贝叶斯滤波器的递归对数后验更新，退出规则在停止时实现无分布的选择性风险且无多重检验惩罚，流状态的向前传递与具有有限精度漂移的前缀重新计算完全等价。ASP在ModelNet40和ModelNet10上分别达到90.62%和93.28%的准确率，在更大骨干网络下比最强脉冲基线低1.7个点，但添加了基线不具备的经认证的任何时刻接口。该机制可原封不动地迁移到密集预测，在ShapeNetPart上实现83.21的实例mIoU，在S3DIS Area 5上实现48.50的mIoU，据我们所知这是S3DIS Area 5上的首个脉冲结果；若将块选择替换为注视点选择，还可迁移到凹形非脉冲Transformer，因此该策略不局限于脉冲骨干网络：成本与观察次数完全线性相关，阈值是可调节的计算旋钮，可将能耗降低2.8倍至1.35倍。一个具体局限是，在我们使用的裁剪尺寸下，一个S3DIS类别无法识别，我们给出了可修复该问题的预测。
英文摘要
Spiking point cloud networks usually scan space in a fixed, input-agnostic order, which leaves the most distinctive resource of spiking computation, the temporal evolution of the membrane potential, unused as a locus of decision-making. Active Spiking Perception (ASP) recasts 3D recognition as an iterative decision process in which the network's own leaky integrate-and-fire (LIF) membrane potential, read as a running belief over the class, selects the next chunk to observe and triggers confidence-margin early exit. A lightweight Slice-Selection Policy scores unvisited farthest-point-sampled chunks from the membrane state and precomputed geometric descriptors, trains end-to-end through a straight-through Gumbel-Softmax, reduces to an argmax at inference, and adds about 2% of backbone parameters. We prove that leaky integration is the recursive log-posterior update of a Bayesian filter, that the exit rule attains distribution-free selective risk with no multiple-testing penalty at the stopping time, and that streaming state carry-forward is exactly equivalent to prefix recomputation with bounded finite-precision drift. ASP reaches 90.62% and 93.28% on ModelNet40 and ModelNet10, 1.7 points below the strongest spiking baseline at a larger backbone, while adding a certified anytime interface no baseline offers. The mechanism transfers unchanged to dense prediction, giving 83.21 instance mIoU on ShapeNetPart and 48.50 mIoU on S3DIS Area 5, to our knowledge the first spiking results on S3DIS Area 5, and, fixation replacing chunk selection, to a foveated non-spiking transformer, so the policy is not tied to spiking backbones: cost is exactly linear in observations and the threshold is a measured compute dial spanning 2.8x to 1.35x less energy. One limitation is concrete: one S3DIS class is unidentifiable at the crop size we use, and we give the prediction that would fix it.
Comments28 pages, 9 figures, 17 tables. Supplementary material included as appendices A-J
