论文标题
解开架构和光流的培训
Disentangling Architecture and Training for Optical Flow
论文作者
论文摘要
培训细节和数据集对筏（如筏）等最新的光流模型有多重要？它们会概括吗？为了探索这些问题，而不是开发新的模型，我们将重新访问三个突出的模型，即PWC-NET，IRR-PWC和RAFT，并采用一组常见的现代培训技术和数据集，并观察到显着的绩效提高，证明了这些培训细节的重要性和一般性。我们新训练的PWC-NET和IRR-PWC模型显示出惊人的改进，与Sintel和Kitti 2015基准相比，最高30％的结果与原始发布的结果相比。他们的表现超过了2015年Kitti上最新的Flow1d，而推断期间的速度快3倍。我们新训练的筏子在2015年的Kitti上取得了4.31％的成绩，比写作时所有已发表的光流方法更准确。我们的结果表明，分析光流方法的性能提高时，分开模型，训练技术和数据集的贡献的好处。我们的源代码将公开可用。
How important are training details and datasets to recent optical flow models like RAFT? And do they generalize? To explore these questions, rather than develop a new model, we revisit three prominent models, PWC-Net, IRR-PWC and RAFT, with a common set of modern training techniques and datasets, and observe significant performance gains, demonstrating the importance and generality of these training details. Our newly trained PWC-Net and IRR-PWC models show surprisingly large improvements, up to 30% versus original published results on Sintel and KITTI 2015 benchmarks. They outperform the more recent Flow1D on KITTI 2015 while being 3x faster during inference. Our newly trained RAFT achieves an Fl-all score of 4.31% on KITTI 2015, more accurate than all published optical flow methods at the time of writing. Our results demonstrate the benefits of separating the contributions of models, training techniques and datasets when analyzing performance gains of optical flow methods. Our source code will be publicly available.
