mlsanityData to 14 Oct 2026

Event-based Optical Flow on Neuromorphic Processor: ANN vs. SNN Comparison based on Activation Sparsification

arXiv 2407.20421 ↗First posted 29 Jul 2024cs.NE cs.CV cs.LGcited by 7

Not on the map. Why?

Abstract (arXiv)

Spiking neural networks (SNNs) for event-based optical flow are claimed to be computationally more efficient than their artificial neural networks (ANNs) counterparts, but a fair comparison is missing in the literature. In this work, we propose an event-based optical flow solution based on activation sparsification and a neuromorphic processor, SENECA. SENECA has an event-driven processing mechanism that can exploit the sparsity in ANN activations and SNN spikes to accelerate the inference of both types of neural networks. The ANN and the SNN for comparison have similar low activation/spike density (~5%) thanks to our novel sparsification-aware training. In the hardware-in-loop experiments designed to deduce the average time and energy consumption, the SNN consumes 44.9ms and 927.0 microjoules, which are 62.5% and 75.2% of the ANN's consumption, respectively. We find that SNN's higher efficiency attributes to its lower pixel-wise spike density (43.5% vs. 66.5%) that requires fewer memory access operations for neuron states.

Key facts. Cited by 7 papers in the map (data through 14 Oct 2026). Builds on 24 of its 45 references that are in the map. First posted on arXiv 29 Jul 2024.

Builds on

24 of 45 references.

References · 24

  1. 2021Sparsity in Deep Learning: Pruning and growth for efficient inference and training in neural networkscited by 527
  2. 2018Unsupervised Event-based Learning of Optical Flow, Depth, and Egomotioncited by 307
  3. 2018The Multi Vehicle Stereo Event Camera Dataset: An Event Camera Dataset for 3D Perceptioncited by 254
  4. 2018Going Deeper in Spiking Neural Networks: VGG and Residual Architecturescited by 245
  5. 2018Direct Training for Spiking Neural Networks: Faster, Larger, Bettercited by 233
  6. 2018EV-FlowNet: Self-Supervised Optical Flow Estimation for Event-based Camerascited by 224
  7. 2019MLPerf Inference Benchmarkcited by 199
  8. 2021E-RAFT: Dense Optical Flow from Event Camerascited by 152
  9. 2017SCNN: An Accelerator for Compressed-sparse Convolutional Neural Networkscited by 135
  10. 2022The Lazy Neuron Phenomenon: On Emergence of Activation Sparsity in Transformerscited by 122
  11. 2021Self-Supervised Learning of Event-Based Optical Flow with Spiking Neural Networkscited by 99
  12. 2020Spike-FlowNet: Event-based Optical Flow Estimation with Energy-Efficient Hybrid Neural Networkscited by 98
12 more
  1. 2022Secrets of Event-Based Optical Flowcited by 92
  2. 2019Focus Is All You Need: Loss Functions For Event-based Visioncited by 83
  3. 2023Taming Contrast Maximization for Learning Sequential, Low-latency, Event-based Optical Flowcited by 39
  4. 2018Accelerating Convolutional Neural Networks via Activation Map Compressioncited by 34
  5. 2023Optical flow estimation from event-based cameras and spiking neural networkscited by 33
  6. 2023Fully neuromorphic vision and control for autonomous drone flightcited by 30
  7. 2022Adaptive-SpikeNet: Event-based Optical Flow Estimation using Spiking Neural Networks with Learnable Neuronal Dynamicscited by 19
  8. 2022Event-based Temporally Dense Optical Flow Estimation with Sequential Learningcited by 15
  9. 2022DeFiNES: Enabling Fast Exploration of the Depth-first Scheduling Space for DNN Accelerators through Analytical Modelingcited by 12
  10. 2023Neuromorphic Optical Flow and Real-time Implementation with Event Camerascited by 11
  11. 2023Accelerating Deep Neural Networks via Semi-Structured Activation Sparsitycited by 10
  12. 2023Open the box of digital neuromorphic processor: Towards effective algorithm-hardware co-designcited by 7

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2024: 1; 2025: 4; 2026: 2.

7 papers.

  1. 2024Learning Normal Flow Directly From Event Neighborhoodscited by 5
  2. 2025Sparse Convolutional Recurrent Learning for Efficient Event-based Neuromorphic Object Detectioncited by 3
  3. 2025Context-aware Sparse Spatiotemporal Learning for Event-based Visioncited by 1
  4. 2025ST-FlowNet: An Efficient Spiking Neural Network for Event-Based Optical Flow Estimationcited by 0
  5. 2025A Scalable, Causal, and Energy Efficient Framework for Neural Decoding with Spiking Neural Networkscited by 0
  6. 2026Spiking and Event-driven Neuromorphic Mamba Models for Efficient Speech Recognitioncited by 0
  7. 2026Towards an Expressivity-Normalized Energy-Demand Comparison of ANNs and SNNscited by 0