Research map / Architectures and efficiency / Spiking networks
SpikingIR: A Novel Converted Spiking Neural Network for Efficient Image Restoration
Yang Ouyang, Zihan Cheng, Xiaotong Luo, Guoqi Li, Yanyun Qu
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add surrogate-gradient fine-tuning to converted SNN (ablation) | SpikingIR on AOD-Net · TestSet A dehazing (T=10) | PSNR: 17.73 -> 18.26 | Improved |
| add CPM convolution layer (ablation) | SpikingIR on AOD-Net · TestSet A dehazing (T=10) | PSNR: 18.26 -> 19.27 | Improved |
| add RMPIS residual membrane potential shortcut (ablation) | SpikingIR on AOD-Net · TestSet A dehazing (T=10) | PSNR: 19.27 -> 19.59 | Improved |
| add MPRN neurons in intermediate layers (ablation) | SpikingIR on AOD-Net · TestSet A dehazing (T=10) | PSNR: 19.59 -> 19.90 | Improved |
| use CPM instead of directly outputting floating-point features (ablation) | SpikingIR on AOD-Net · TestSet A dehazing (T=10) | PSNR: 19.23 -> 19.59 | Improved |
| SpikingIR vs prior ANN-to-SNN conversion (Lan et al.) with fewer timesteps | VDSR x4 super-resolution · Set5 x4 | PSNR: 28.722 -> 31.264 | Improved |
| convert dehazing ANN with SpikingIR | AOD-Net · SOTS-outdoor | PSNR: 20.12 -> 21.47 | Improved |
| convert denoising ANN with SpikingIR | DnCNN · Set12 sigma=15 | PSNR: 32.859 -> 32.758 | Hurt |
| convert VDSR with SpikingIR | VDSR · Set5 x2 | PSNR: 37.653 -> 37.459 | Hurt |
| use SpikingIR instead of ANN for energy | AOD-Net · TestSet A dehazing | energy (mJ): 2.289 -> 0.931 (T=5) | Improved |
| increase timesteps from 10 to 15 | SpikingIR on DnCNN · Set12 sigma=25 | PSNR: 30.160 -> 30.285 | Improved |
Closest papers by meaning
- Adaptive Calibration: A Unified Conversion Framework of Spiking Neural Networks (AAAI 2025 · similarity 0.88)
- Spiking Meets Attention: Efficient Remote Sensing Image Super-Resolution with Attention Spiking Neural Networks (NeurIPS 2025 · similarity 0.87)
- Spike2Former: Efficient Spiking Transformer for High-performance Image Segmentation (AAAI 2025 · similarity 0.87)
- Rethinking the Membrane Dynamics and Optimization Objectives of Spiking Neural Networks (NeurIPS 2024 · similarity 0.87)
- Memory-Efficient Reversible Spiking Neural Networks (AAAI 2024 · similarity 0.87)
- Inference-Scale Complexity in ANN-SNN Conversion for High-Performance and Low-Power Applications (CVPR 2025 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
