Razor SNN: Efficient Spiking Neural Network with Temporal Embeddings
This work addresses efficiency issues in neuromorphic computing for event-based vision tasks, representing an incremental improvement in SNN methods.
The paper tackles the inefficiency of spiking neural networks (SNNs) in processing dense temporal event streams from dynamic vision sensors by proposing Razor SNN, which prunes redundant event frames using temporal embeddings, achieving competitive performance on four benchmarks including DVS 128 Gesture and CIFAR10-DVS.
The event streams generated by dynamic vision sensors (DVS) are sparse and non-uniform in the spatial domain, while still dense and redundant in the temporal domain. Although spiking neural network (SNN), the event-driven neuromorphic model, has the potential to extract spatio-temporal features from the event streams, it is not effective and efficient. Based on the above, we propose an events sparsification spiking framework dubbed as Razor SNN, pruning pointless event frames progressively. Concretely, we extend the dynamic mechanism based on the global temporal embeddings, reconstruct the features, and emphasize the events effect adaptively at the training stage. During the inference stage, eliminate fruitless frames hierarchically according to a binary mask generated by the trained temporal embeddings. Comprehensive experiments demonstrate that our Razor SNN achieves competitive performance consistently on four events-based benchmarks: DVS 128 Gesture, N-Caltech 101, CIFAR10-DVS and SHD.