将图像升维处理，用3D网络实现高效高精度的密集预测
Spatial Lifting for Dense Prediction
- 把2D图像升维到更高维度，用3D网络处理
- 参数量减少超98%，推理成本更低，性能仍具竞争力
- 自带结构化输出，便于训练监督和测试评估
我们提出空间提升（Spatial Lifting, SL），一种面向密集预测任务的新方法。SL将标准输入（如2D图像）升维至高维空间，并使用为该维度设计的网络（如3D U-Net）进行处理。反直觉的是，该方法在基准任务上表现优异，同时显著降低推理成本并大幅减少模型参数量。SL框架在升维方向产生内在结构化输出，这种结构在训练时支持密集监督，在测试时可实现近乎零成本的预测质量评估。我们在19个基准数据集（13个语义分割、6个深度估计）上验证了该方法，相比传统方法，参数量减少超过98%（以U-Net为例），推理成本降低，且保持了竞争性性能。空间提升引入了一种新的视觉建模范式，为实现更高效、准确和可靠的密集预测深度网络提供了新路径。
原文摘要 · Abstract (English)
We present Spatial Lifting (SL), a novel methodology for dense prediction tasks. SL operates by lifting standard inputs, such as 2D images, into a higher-dimensional space and subsequently processing them using networks designed for that higher dimension, such as a 3D U-Net. Counterintuitively, this dimensionality lifting allows us to achieve good performance on benchmark tasks compared to conventional approaches, while reducing inference costs and significantly lowering the number of model parameters. The SL framework produces intrinsically structured outputs along the lifted dimension. This emergent structure facilitates dense supervision during training and enables robust, near-zero-additional-cost prediction quality assessment at test time. We validate our approach across 19 benchmark datasets (13 for semantic segmentation and 6 for depth estimation), demonstrating competitive dense prediction performance while reducing the model parameter count by over 98% (in the U-Net case) and lowering inference costs. Spatial Lifting introduces a new vision modeling paradigm that offers a promising path toward more efficient, accurate, and reliable deep networks for dense prediction tasks in vision.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。