Research map / Video / Video segmentation and tracking
Event-Based Motion Deblurring Using Task-Oriented 3D Gaussian Event Representations
Shengdong Xue, Haoxiang Ma, Hao Chen, Zhen Yang, Yongjian Deng
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 |
|---|---|---|---|
| replace Voxel Grid with learnable 3D Gaussian event representation (ablation) | deblurring network · GoPro | PSNR (dB): 36.13 -> 36.51 | Improved |
| add blur-map prior (ablation) | 3D-GSER network · GoPro | PSNR (dB): 36.51 -> 36.61 | Improved |
| add bidirectional attention fusion (BAF) (ablation) | 3D-GSER network · GoPro | PSNR (dB): 36.51 -> 36.76 | Improved |
| combine blur map and BAF (ablation) | 3D-GSER network · GoPro | PSNR (dB): 36.51 -> 36.86 | Improved |
| add full modules on REBlur (ablation) | 3D-GSER network · REBlur | PSNR (dB): 38.01 -> 38.68 | Improved |
| use learning-based EST instead of Voxel Grid (ablation) | deblurring network · GoPro | PSNR (dB): 36.13 -> 35.86 | Hurt |
| use handcrafted SCER instead of Voxel Grid (ablation) | deblurring network · GoPro | PSNR (dB): 36.13 -> 35.95 | Hurt |
| proposed method vs best prior on HS-ERGB | full method · HS-ERGB | PSNR (dB): 27.93 -> 28.55 | Improved |
| proposed method vs best prior on GoPro (SSIM) | full method · GoPro | SSIM: 0.979 (EIFNet) vs 0.977 | Mixed |
Closest papers by meaning
- Learning Scale-Aware Spatio-temporal Implicit Representation for Event-based Motion Deblurring (ICML 2024 · similarity 0.92)
- Motion-adaptive Transformer for Event-based Image Deblurring (AAAI 2025 · similarity 0.92)
- Motion Deblurring via Spatial-Temporal Collaboration of Frames and Events (AAAI 2024 · similarity 0.91)
- Frequency-aware Event-based Video Deblurring for Real-World Motion Blur (CVPR 2024 · similarity 0.91)
- DiET-GS: Diffusion Prior and Event Stream-Assisted Motion Deblurring 3D Gaussian Splatting (CVPR 2025 · similarity 0.89)
- Diffusion-based Event Generation for High-Quality Image Deblurring (CVPR 2025 · similarity 0.89)
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.
