Research map / Video / Video segmentation and tracking
Enhancing Motion Deblurring in High-Speed Scenes with Spike Streams
Shiyan Chen, Jiyuan Zhang, Yajing Zheng, Tiejun Huang, Zhaofei Yu
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 |
|---|---|---|---|
| remove CAMMA branch (ablation) | SpkDeblurNet · Spk-X4K1000FPS e=65 | PSNR: 35.94 -> 35.64 | Hurt |
| remove CAMMA branch (ablation) | SpkDeblurNet · Spk-X4K1000FPS e=33 | PSNR: 37.42 -> 37.20 | Hurt |
| replace TCAF with concatenation (ablation) | SpkDeblurNet · Spk-X4K1000FPS e=33 | PSNR: 37.42 -> 37.36 | Hurt |
| replace TCAF with concatenation (ablation) | SpkDeblurNet · Spk-X4K1000FPS e=65 | PSNR: 35.94 -> 35.81 | Hurt |
| replace spike input representation with CSFI (ablation) | SpkDeblurNet · Spk-X4K1000FPS e=65 | PSNR: 35.94 -> 34.22 | Hurt |
| replace spike input representation with cumulated spikes (ablation) | SpkDeblurNet · Spk-X4K1000FPS e=65 | PSNR: 35.94 -> 34.92 | Hurt |
| add CAMMA to spike reconstruction with CSFI input (ablation) | spike reconstruction branch · Spk-X4K1000FPS e=33 | PSNR: 33.67 -> 37.25 | Improved |
| add CAMMA to spike reconstruction with cumulated spike input (ablation) | spike reconstruction branch · Spk-X4K1000FPS e=33 | PSNR: 35.58 -> 37.54 | Improved |
| add CAMMA to spike reconstruction with CSFI input (ablation) | spike reconstruction branch · Spk-X4K1000FPS e=65 | PSNR: 33.69 -> 36.13 | Improved |
| add CAMMA with spike stream input (ablation) | spike reconstruction branch · Spk-X4K1000FPS | PSNR: +0.4-0.7 dB | Improved |
| use spike streams with bidirectional fusion instead of event-based EFNet | EFNet · Spk-X4K1000FPS e=65 | PSNR: 33.53 -> 35.94 | Improved |
| use spike streams instead of image-only NAFNet | NAFNet · Spk-X4K1000FPS e=65 | PSNR: 29.06 -> 35.94 | Improved |
Closest papers by meaning
- SpikeReveal: Unlocking Temporal Sequences from Real Blurry Inputs with Spike Streams (NeurIPS 2024 · similarity 0.93)
- Spike-guided Motion Deblurring with Unknown Modal Spatiotemporal Alignment (CVPR 2024 · similarity 0.92)
- DeblurSR: Event-Based Motion Deblurring under the Spiking Representation (AAAI 2024 · similarity 0.89)
- Transient Glimpses: Unveiling Occluded Backgrounds through the Spike Camera (AAAI 2024 · similarity 0.88)
- Rethinking High-speed Image Reconstruction Framework with Spike Camera (AAAI 2025 · similarity 0.88)
- Motion-adaptive Transformer for Event-based Image Deblurring (AAAI 2025 · similarity 0.88)
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.
