Research map / Video / Video segmentation and tracking
HATIR: Heat-Aware Diffusion for Turbulent Infrared Video Super-Resolution
Yang Zou, Xingyue Zhu, Kaiqi Han, Jun Ma, Xingyuan Li, Zhiying Jiang, Jinyuan Liu
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 |
|---|---|---|---|
| use PhasorFlow instead of SpyNet for flow (ablation) | HATIR · ablation test set | PSNR: 28.9387 -> 33.6507 | Improved |
| use PhasorFlow instead of SpyNet for flow (ablation) | HATIR · ablation test set | VMAF: 33.1466 -> 45.3972 | Improved |
| add IR-SAA to the decoder (with TMG) (ablation) | TAD · ablation test set | PSNR: 28.4125 -> 32.2391 | Improved |
| add TMG to the decoder (with IR-SAA) (ablation) | TAD · ablation test set | PSNR: 27.3985 -> 32.2391 | Improved |
| add both IR-SAA and TMG to the decoder (ablation) | TAD · ablation test set | LPIPS: 0.2862 -> 0.1358 | Improved |
| add Mphasor to occlusion mask in guidance (ablation) | heat-aware guidance · ablation test set | PSNR: 31.6965 -> 32.1595 | Improved |
| add Mocc to Mphasor in guidance (ablation) | heat-aware guidance · ablation test set | PSNR: 28.9239 -> 32.1595 | Improved |
| use Mphasor mask only versus no mask (ablation) | heat-aware guidance · ablation test set | VMAF: 34.0477 -> 30.4051 | Hurt |
| add heat-aware masks versus none (ablation) | heat-aware guidance · ablation test set | PSNR: 26.3073 -> 32.1595 | Improved |
| HATIR vs DATUM on small M3FD Set5 | HATIR · M3FD Set5 | VMAF: 45.8987 -> 44.6731 | Hurt |
| HATIR vs MGLDVSR on M3FD Set5 | HATIR · M3FD Set5 | DISTS: 0.1513 -> 0.1576 | Hurt |
| HATIR vs best baseline across camera motions | HATIR · FLIR-IVSR | PSNR: 29.6184 -> 33.3719 | Improved |
Closest papers by meaning
- Thermal Diffusion Matters: Infrared Spatial-Temporal Video Super-Resolution through Heat Conduction Priors (CVPR 2026 · similarity 0.89)
- Toward Real-world Infrared Image Super-Resolution: A Unified Autoregressive Framework and Benchmark Dataset (CVPR 2026 · similarity 0.88)
- DifIISR: A Diffusion Model with Gradient Guidance for Infrared Image Super-Resolution (CVPR 2025 · similarity 0.88)
- Turb-Seg-Res: A Segment-then-Restore Pipeline for Dynamic Videos with Atmospheric Turbulence (CVPR 2024 · similarity 0.85)
- One-Step Diffusion for Detail-Rich and Temporally Consistent Video Super-Resolution (NeurIPS 2025 · similarity 0.85)
- One Diffusion Step to Real-World Super-Resolution via Flow Trajectory Distillation (ICML 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.
