Research map / Video / Video segmentation and tracking
Deep SR-ITM: Joint Learning of Super-Resolution and Inverse Tone-Mapping for 4K UHD HDR Applications
Soo Ye Kim, Jihyong Oh, Munchurl Kim
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 guided filter base/detail decomposition of the input (ablation) | Deep SR-ITM · Ablation of Deep SR-ITM (SF=2) | PSNR (dB): 35.29 -> 35.37 | Improved |
| add modulation blocks (ablation) | Deep SR-ITM · Ablation of Deep SR-ITM (SF=2) | PSNR (dB): 35.37 -> 35.59 | Improved |
| add skip connections on top of modulation (ablation) | Deep SR-ITM · Ablation of Deep SR-ITM (SF=2) | PSNR (dB): 35.59 -> 35.58 | No clear effect |
| decompose input and stack image, base and detail in a single pass (ablation) | toy network · Toy network, input decomposition | PSNR (dB): 38.11 -> 38.25 | Improved |
| stack the LR SDR image with each of two decomposed passes (ablation) | toy network · Toy network, input decomposition | PSNR (dB): 38.11 -> 38.46 | Improved |
| use separate passes for base and detail instead of one stacked pass (ablation) | toy network · Toy network, input decomposition | PSNR (dB): 38.25 -> 38.46 | Improved |
| add a third pass for the raw image instead of stacking it (ablation) | toy network · Toy network, input decomposition | PSNR (dB): 38.46 -> 38.21 | Hurt |
| concatenate image with base layer as modulation input (ablation) | toy network · Toy network, modulation input | PSNR (dB): 38.43 -> 38.52 | Improved |
| use Deep SR-ITM instead of multi-purpose CNN (SF=2) | Deep SR-ITM · 4K HDR test set (SF=2) | PSNR (dB): 34.11 -> 35.58 | Improved |
| use Deep SR-ITM instead of multi-purpose CNN (SF=4) | Deep SR-ITM · 4K HDR test set (SF=4) | PSNR (dB): 33.10 -> 33.61 | Improved |
| use Deep SR-ITM instead of multi-purpose CNN on HDR-VDP (SF=4) | Deep SR-ITM · 4K HDR test set (SF=4) | HDR-VDP (Q): 56.41 -> 56.07 | Hurt |
Later papers that compared with this method
2 later accepted papers report Deep SR-ITM as a baseline in their results tables; 2 of them report a better number on most of the results they share with it.
- A New Journey From SDRTV to HDRTV (ICCV 2021 · better on 2 of 2 shared results)
- JSI-GAN: GAN-Based Joint Super-Resolution and Inverse Tone-Mapping with Pixel-Wise Task-Specific Filters for UHD HDR Video (AAAI 2020 · better on 4 of 4 shared results)
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- JSI-GAN: GAN-Based Joint Super-Resolution and Inverse Tone-Mapping with Pixel-Wise Task-Specific Filters for UHD HDR Video (AAAI 2020 · similarity 0.90)
- Ultra-High-Definition Image HDR Reconstruction via Collaborative Bilateral Learning (ICCV 2021 · similarity 0.86)
- Deep Video Inverse Tone Mapping Based on Temporal Clues (CVPR 2024 · similarity 0.85)
- From Zero to Detail: Deconstructing Ultra-High-Definition Image Restoration from Progressive Spectral Perspective (CVPR 2025 · similarity 0.85)
- Edge-SD-SR: Low Latency and Parameter Efficient On-device Super-Resolution with Stable Diffusion via Bidirectional Conditioning (CVPR 2025 · similarity 0.85)
- Correlation Matching Transformation Transformers for UHD Image Restoration (AAAI 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.
