FastGaMer：面向实用逆色调映射的高效增益图学习
FastGaMer: Efficient GainMap Learning for Practical Inverse Tone Mapping
把 HDR 重建从直接回归像素值改为重建逐通道颜色增益图，并用双边网格与可学习 3D LUT 分别反转局部/全局退化，这一「镜像退化过程」的分解思路可迁移到其他需要恢复被非线性压缩信息的复原任务（如低光照增强、压缩伪影去除）。
原文依据 Instead of regressing HDR values, we reconstruct a color gain map, which preserves per-channel amplification
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论当 SDR 输入由局部色调映射产生时，逆色调映射（ITM）会变得显著困难：局部色调映射同时施加全局辐射压缩与空间变化的适应，从而扭曲动态范围、对比度以及逐通道颜色比例。现有 ITM 方法忽略了这种退化结构，要么直接回归 HDR 值，要么依赖单通道增益图，而后者只缩放亮度，无法恢复被压缩的动态范围与广色域。我们提出 FastGaMer，一个结构化且与分辨率无关的 ITM 框架，显式地镜像这一退化过程。我们不再回归 HDR 值，而是重建颜色增益图，它保留逐通道放大、简化学习并支持恰当的色域扩展。局部与全局退化分别通过动态双边网格和可学习 3D LUT 加以反转，随后由一个轻量神经调制器进行全局细化与一致性处理。所有高分辨率操作均无需网络，从而带来出色的效率。为在真实局部 TMO 退化下支持颜色增益图监督，我们构建了一个包含 8000 多对 4K SDR-GM 的数据集，并额外提供一个真实拍摄的测试集。FastGaMer 相比此前的轻量级 ITM 方法提升 +1.4 dB PQ-PSNR，运行时间降低 70%，处理 4K 图像仅需 6.2 ms，兼顾高精度与实时性能。
英文原文摘要 1410 字符
Inverse tone mapping (ITM) becomes significantly harder when the SDR input is produced by local tone mapping, which jointly applies global radiometric compression and spatially varying adaptations that distort dynamic range, contrast, and channel-wise color ratios. Existing ITM methods ignore this degradation structure and either regress HDR values directly or rely on a single-channel gain map, which scale luminance only and cannot restore the compressed dynamic range and wide color gamut.We introduce FastGaMer, a structured and resolution-agnostic ITM framework that explicitly mirrors this degradation process. Instead of regressing HDR values, we reconstruct a color gain map, which preserves per-channel amplification, simplifies learning, and enables proper gamut extension. Local and global degradations are inverted separately using dynamic bilateral grids and learnable 3D LUTs, followed by a lightweight neural modulator for global refinement and coherence. All high-resolution operations are network-free, yielding exceptional efficiency.To support color-GM supervision under realistic local TMO degradations, we create a dataset of over 8,000 4K SDR-GM pairs with an additional real-captured test set. FastGaMer outperforms prior lightweight ITM methods by +1.4 dB PQ-PSNR, reduces runtime by 70%, and processes 4K images in only 6.2 ms, achieving both high accuracy and real-time performance.