Introduces the first large-scale event-based HAR benchmark under low-light and 6-DoF shaking conditions with IMU data, and an EIS-HAR pipeline using non-linear warping for motion compensation plus a four-stage hybrid network that outperforms prior methods on three datasets.
Optimal ann- snn conversion for high-accuracy and ultra-low-latency spiking neural networks,
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
representative citing papers
Spikinghash combines 3D-DWT Spiking WaveMixer, Spiking Self-Attention, and a dynamic soft similarity loss to produce energy-efficient hash codes for DVS data retrieval.
ELSA is a near-SRAM dataflow architecture realizing elastic inference in SNNs via fine-grained spine/token pipelines, bundled AER, and mini-batch Gustavson products, delivering up to 3.4x speedup and 22.1x energy gains over SOTA accelerators on ResNet-50.
citing papers explorer
-
DarkShake-DVS: Event-based Human Action Recognition under Low-light andShaking Camera Conditions
Introduces the first large-scale event-based HAR benchmark under low-light and 6-DoF shaking conditions with IMU data, and an EIS-HAR pipeline using non-linear warping for motion compensation plus a four-stage hybrid network that outperforms prior methods on three datasets.
-
Temporal-Aware Spiking Transformer Hashing Based on 3D-DWT
Spikinghash combines 3D-DWT Spiking WaveMixer, Spiking Self-Attention, and a dynamic soft similarity loss to produce energy-efficient hash codes for DVS data retrieval.
-
ELSA: An ELastic SNN Inference Architecture for Efficient Neuromorphic Computing
ELSA is a near-SRAM dataflow architecture realizing elastic inference in SNNs via fine-grained spine/token pipelines, bundled AER, and mini-batch Gustavson products, delivering up to 3.4x speedup and 22.1x energy gains over SOTA accelerators on ResNet-50.