VoDaSuRe：一个揭示体数据超分辨率中域偏移的大规模数据集
VoDaSuRe: A Large-Scale Dataset Revealing Domain Shift in Volumetric Super-Resolution
该论文揭示了体数据超分辨率模型在真实低分辨率扫描上性能被高估的问题，并提供了大规模配对数据集VoDaSuRe，可用于重新评估和开发更鲁棒的SR方法，避免因训练数据下采样而导致的域偏移。
原文依据 we show that much of this performance stems from training on downsampled data rather than real low-resolution scans
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论近期体数据超分辨率（SR）的进展在医学和科学成像中展现出强大性能，基于Transformer和CNN的方法即使在极端缩放因子下也取得了令人印象深刻的结果。在本工作中，我们表明这种性能很大程度上源于在下采样数据而非真实低分辨率扫描上的训练。这种对下采样的依赖部分是由于配对的高分辨率和低分辨率3D数据集的稀缺。为解决这一问题，我们引入了VoDaSuRe，一个包含配对高分辨率和低分辨率扫描的大规模体数据集。在VoDaSuRe上训练模型时，我们揭示了一个显著差异：在下采样数据上训练的SR模型产生的预测比在真实低分辨率扫描上训练的模型清晰得多，后者会平滑精细结构。相反，将在下采样数据上训练的模型应用于真实扫描时，能保留更多结构但不准确。我们的发现表明，当前的SR方法被高估了——当应用于真实数据时，它们无法恢复低分辨率扫描中丢失的结构，而是预测平滑的平均值。我们认为，基于深度学习的体数据SR的进展需要具有高复杂度的配对真实扫描数据集，例如VoDaSuRe。我们的数据集和代码可通过以下网址公开获取：https://augusthoeg.github.io/VoDaSuRe/
英文原文摘要 1357 字符
Recent advances in volumetric super-resolution (SR) have demonstrated strong performance in medical and scientific imaging, with transformer- and CNN-based approaches achieving impressive results even at extreme scaling factors. In this work, we show that much of this performance stems from training on downsampled data rather than real low-resolution scans. This reliance on downsampling is partly driven by the scarcity of paired high- and low-resolution 3D datasets. To address this, we introduce VoDaSuRe, a large-scale volumetric dataset containing paired high- and low-resolution scans. When training models on VoDaSuRe, we reveal a significant discrepancy: SR models trained on downsampled data produce substantially sharper predictions than those trained on real low-resolution scans, which smooth fine structures. Conversely, applying models trained on downsampled data to real scans preserves more structure but is inaccurate. Our findings suggest that current SR methods are overstated - when applied to real data, they do not recover structures lost in low-resolution scans and instead predict a smoothed average. We argue that progress in deep learning-based volumetric SR requires datasets with paired real scans of high complexity, such as VoDaSuRe. Our dataset and code are publicly available through: https://augusthoeg.github.io/VoDaSuRe/