Research map / Image generation / GANs
Deep Recursive HDRI: Inverse Tone Mapping using Generative Adversarial Networks
Siyeong Lee, Gwon Hwan An, Suk-Ju Kang
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 GAN loss to L1 loss (ablation) | Deep Recursive HDRI · VDS, Reinhard TMO | PSNR: 28.12 -> 32.14 | Improved |
| add GAN loss to L1 loss (ablation) | Deep Recursive HDRI · VDS, Kim and Kautz TMO | PSNR: 23.41 -> 25.36 | Improved |
| add GAN loss to L1 loss (ablation) | Deep Recursive HDRI · VDS | HDR-VDP-2 quality score: 55.78 -> 57.28 | Improved |
| add GAN loss to L1 loss (ablation) | Deep Recursive HDRI · HDREye, Reinhard TMO | PSNR: 26.19 -> 26.71 | Improved |
| add GAN loss to L1 loss (ablation) | Deep Recursive HDRI · HDREye, Kim and Kautz TMO | PSNR: 21.69 -> 22.31 | Improved |
| add GAN loss to L1 loss (ablation) | Deep Recursive HDRI · HDREye | HDR-VDP-2 quality score: 49.00 -> 48.85 | No clear effect |
| use recursive GAN-based networks instead of deep chain HDRI | Deep Recursive HDRI · VDS, Reinhard TMO | PSNR: 30.86 -> 32.14 | Improved |
| use recursive GAN-based networks instead of deep chain HDRI | Deep Recursive HDRI · HDREye, Reinhard TMO | PSNR: 25.77 -> 26.71 | Hurt |
| use proposed method instead of Endo et al. | Deep Recursive HDRI · VDS, Reinhard TMO | PSNR: 25.49 -> 32.14 | Improved |
| use proposed method instead of Eilertsen et al. | Deep Recursive HDRI · VDS, Reinhard TMO | PSNR: 17.97 -> 32.14 | Improved |
| use proposed method instead of deep chain HDRI for negative exposure | Deep Recursive HDRI · LDR stack EV -3 | PSNR: 24.33 -> 29.15 | Improved |
| use proposed method instead of deep chain HDRI for positive exposure | Deep Recursive HDRI · LDR stack EV +2 | PSNR: 29.65 -> 29.43 | Hurt |
Closest papers by meaning
- GlowGAN: Unsupervised Learning of HDR Images from LDR Images in the Wild (ICCV 2023 · similarity 0.88)
- Zero-Shot Structure-Preserving Diffusion Model for High Dynamic Range Tone Mapping (CVPR 2024 · similarity 0.88)
- Generating Content for HDR Deghosting from Frequency View (CVPR 2024 · similarity 0.85)
- Learning Differential Pyramid Representation for Tone Mapping (NeurIPS 2025 · similarity 0.85)
- GaussHDR: High Dynamic Range Gaussian Splatting via Learning Unified 3D and 2D Local Tone Mapping (CVPR 2025 · similarity 0.85)
- Self-Supervised High Dynamic Range Imaging with Multi-Exposure Images in Dynamic Scenes (ICLR 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.
