Research map / Vision-language and multimodal / Cross-modal learning
DeLux: Cross-Modal Local Artifact Restoration in Video Using Neuromorphic Data
Bartosz Stachowiak, Dariusz Brzezinski
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 event modality input (ablation) | DeLux · Synthetic artifacts, removal | PSNR: 33.757 -> 36.461 | Improved |
| add event modality input (ablation) | DeLux · Synthetic artifacts, removal | MS-SSIM: 0.986 -> 0.991 | Improved |
| add event modality input (ablation) | DeLux · Real-world recordings | Delta SAS: 67.06% -> 63.22% | Hurt |
| add explicit artifact detector (ablation) | DeLux · Synthetic artifacts, removal | PSNR: 30.822 -> 36.461 | Improved |
| add explicit artifact detector (ablation) | DeLux · Real-world recordings | Delta SAS: -195.34% -> 63.22% | Improved |
| add non-mask loss (ablation) | DeLux · Real-world recordings | Delta SAS: -1703.38% -> 63.22% | Improved |
| add non-mask loss (ablation) | DeLux · Synthetic artifacts, detection | accuracy: 0.971 -> 0.925 | Hurt |
| add non-mask loss (ablation) | DeLux · Synthetic artifacts, detection | F1-score: 0.785 -> 0.613 | Hurt |
| add non-mask loss (ablation) | DeLux · Synthetic artifacts, removal | PSNR: 35.839 -> 36.461 | Improved |
| use cross-modal event-guided DeLux instead of global-tone-mapping event HDR (HDRev-Diff) | DeLux · Real-world recordings | Delta SAS: -1402.87% -> 63.22% | Improved |
| use modular DeLux instead of single-network DAD trained on same data | DeLux · Synthetic artifacts, removal | PSNR: 29.525 -> 36.461 | Improved |
Closest papers by meaning
- Bricker to BRACE: A Bracket Exposure RAW Dataset and Restoration Model for Flicker-Banding (NeurIPS 2026 · similarity 0.85)
- REALM: An RGB and Event Aligned Latent Manifold for Cross-Modal Perception (ECCV 2026 · similarity 0.85)
- Spatio-Temporal Conditional Denoising Transformer for Modality-Missing RGBT Tracking (CVPR 2026 · similarity 0.85)
- Event-Guided Consistent Video Enhancement with Modality-Adaptive Diffusion Pipeline (NeurIPS 2025 · similarity 0.85)
- BurstDeflicker: A Benchmark Dataset for Flicker Removal in Dynamic Scenes (NeurIPS 2025 · similarity 0.84)
- UniLumos: Fast and Unified Image and Video Relighting with Physics-Plausible Feedback (NeurIPS 2025 · similarity 0.83)
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.
