arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1902.09145cs.CV
收藏

DDFlow:利用无标记数据蒸馏学习光流

DDFlow: Learning Optical Flow with Unlabeled Data Distillation

The Chinese University of Hong Kong(香港中文大学);Tencent AI Lab(腾讯人工智能实验室)

Pengpeng Liu, Irwin King, Michael R. Lyu, Jia Xu

摘要:

我们提出了DDFlow,一个数据蒸馏方法来学习光流估计从未标记的数据。该方法从教师网络中提取可靠的预测,并使用这些预测作为注释来引导学生网络学习光流。与现有的工作依赖于手工制作的能量项来处理遮挡,我们的方法是数据驱动的,并学习被遮挡像素的光流。这使我们能够用更简单的损失函数来训练我们的模型,并实现更高的精度。我们对具有挑战性的Flying Chairs,MPI Sintel,KITTI 2012和2015基准进行了严格的评估,并表明我们的方法在实时的运行时显著优于所有现有的无监督学习方法。

英文摘要:

We present DDFlow, a data distillation approach to learning optical flow estimation from unlabeled data. The approach distills reliable predictions from a teacher network, and uses these predictions as annotations to guide a student network to learn optical flow. Unlike existing work relying on hand-crafted energy terms to handle occlusion, our approach is data-driven, and learns optical flow for occluded pixels. This enables us to train our model with a much simpler loss function, and achieve a much higher accuracy. We conduct a rigorous evaluation on the challenging Flying Chairs, MPI Sintel, KITTI 2012 and 2015 benchmarks, and show that our approach significantly outperforms all existing unsupervised learning methods, while running at real time.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑