Research map / Label-efficient and robust learning / Self-supervised and contrastive
Towards Real-World HDRTV Reconstruction: A Data Synthesis-Based Approach
Zhen Cheng, Tao Wang, Yong Li, Fenglong Song, Chang Chen, Zhiwei Xiong
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 HTMP content loss to adversarial loss (ablation) | data synthesis network · RealHDRTV synthesis | PSNR: 11.74 -> 24.54 | Improved |
| add adversarial loss to HTMP loss (ablation) | data synthesis network · RealHDRTV synthesis | PSNR: 24.24 -> 24.54 | Improved |
| use region-aware HTMP loss instead of L1 to Linear TMO (ablation) | data synthesis network · RealHDRTV synthesis | PSNR: 16.35 -> 24.54 | Improved |
| use region-aware HTMP loss instead of L1 to mu-law TMO (ablation) | data synthesis network · RealHDRTV synthesis | PSNR: 18.18 -> 24.54 | Improved |
| use region-aware HTMP loss instead of L1 to Youtube TMO (ablation) | data synthesis network · RealHDRTV synthesis | PSNR: 23.32 -> 24.54 | Improved |
| add condition network with tone mapping results (ablation) | data synthesis network · RealHDRTV synthesis | PSNR: 24.08 -> 24.54 | Improved |
| condition on prior TMO results instead of the input HDRTV itself (ablation) | data synthesis network · RealHDRTV synthesis | PSNR: 24.15 -> 24.54 | Improved |
| add local adjustment stream to global stream (ablation) | data synthesis network · RealHDRTV synthesis | PSNR: 24.33 -> 24.54 | Improved |
| add global mapping stream to local stream (ablation) | data synthesis network · RealHDRTV synthesis | PSNR: 24.44 -> 24.54 | Improved |
| train HDRTVNet-AGCM on data synthesized by learned method instead of Hable | HDRTVNet-AGCM · RealHDRTV reconstruction | PSNR: 25.44 -> 28.04 | Improved |
| synthesize SDRTV with learned method instead of Hable | SDRTV synthesis · RealHDRTV synthesis | PSNR: 23.27 -> 24.54 | Improved |
| use TMOs with high TMQI (Rank) for synthesis | TMO synthesis · RealHDRTV synthesis | PSNR: 22.99 -> 16.57 | Hurt |
Closest papers by meaning
- Towards Real-World HDR Video Reconstruction: A Large-Scale Benchmark Dataset and A Two-Stage Alignment Network (CVPR 2024 · similarity 0.86)
- GaussHDR: High Dynamic Range Gaussian Splatting via Learning Unified 3D and 2D Local Tone Mapping (CVPR 2025 · similarity 0.85)
- Zero-Shot Structure-Preserving Diffusion Model for High Dynamic Range Tone Mapping (CVPR 2024 · similarity 0.85)
- High Dynamic Range Video Compression: A Large-Scale Benchmark Dataset and A Learned Bit-depth Scalable Compression Algorithm (CVPR 2025 · similarity 0.85)
- Learning Differential Pyramid Representation for Tone Mapping (NeurIPS 2025 · similarity 0.84)
- Generating Content for HDR Deghosting from Frequency View (CVPR 2024 · 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.
