Research map / Architectures and efficiency / CNN design
Image Dehazing Transformer With Transmission-Aware 3D Position Embedding
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 transmission-aware 3D position embedding (ablation) | DeHamer · NH-HAZE | PSNR (dB): 18.90 -> 20.66 | Helped |
| add the Transformer module (ablation) | DeHamer · NH-HAZE | PSNR (dB): 18.25 -> 20.66 | Helped |
| add the CNN encoder module (ablation) | DeHamer · NH-HAZE | PSNR (dB): 16.31 -> 20.66 | Helped |
| add the pyramid pooling module to the CNN encoder (ablation) | DeHamer · NH-HAZE | PSNR (dB): 19.04 -> 20.66 | Helped |
| use feature modulation instead of feature addition (ablation) | DeHamer · NH-HAZE | PSNR (dB): 17.07 -> 20.66 | Helped |
| use feature modulation instead of feature concatenation (ablation) | DeHamer · NH-HAZE | PSNR (dB): 18.69 -> 20.66 | Helped |
| add the multiscale residual block to the CNN decoder (ablation) | DeHamer · NH-HAZE | PSNR (dB): 18.68 -> 20.66 | Helped |
| use DeHamer rather than the strongest listed baseline | image dehazing · SOTS-Indoor | SSIM: 0.9886 -> 0.9881 | Hurt |
| use DeHamer rather than the strongest listed baseline | image dehazing · SOTS-Outdoor | PSNR (dB): 34.81 -> 35.18 | Helped |
| use DeHamer rather than the strongest listed baseline | image dehazing · Dense-Haze | PSNR (dB): 15.13 -> 16.62 | Helped |
| use DeHamer rather than the strongest listed baseline | image dehazing · NH-HAZE | PSNR (dB): 18.33 -> 20.66 | Helped |
Closest papers by meaning
- SelfPromer: Self-Prompt Dehazing Transformers with Depth-Consistency (AAAI 2024 · similarity 0.86)
- Prior-guided Hierarchical Harmonization Network for Efficient Image Dehazing (AAAI 2025 · similarity 0.85)
- PromptHaze: Prompting Real-world Dehazing via Depth Anything Model (AAAI 2025 · similarity 0.85)
- Exploiting Diffusion Prior for Real-World Image Dehazing with Unpaired Training (AAAI 2025 · similarity 0.85)
- Tokenize Image Patches: Global Context Fusion for Effective Haze Removal in Large Images (CVPR 2025 · similarity 0.84)
- Efficient Visual State Space Model for Image Deblurring (CVPR 2025 · similarity 0.84)
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.