arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1807.11037cs.CV
收藏

视频语义分割的有效不确定性估计

Efficient Uncertainty Estimation for Semantic Segmentation in Videos

National Tsing Hua University(国立清华大学);Umbo Computer Vision(Umbo计算机视觉)

Po-Yu Huang, Wan-Ting Hsu, Chun-Yueh Chiu, Ting-Fan Wu, Min Sun

摘要:

深度学习中的不确定性估计最近变得更加重要。如果我们不知道一个深度学习模型对决策是否确定,它就无法应用于实际应用中。一些文献提出了贝叶斯神经网络,它可以估计的不确定性,通过蒙特卡洛辍学(MC辍学)。但是,MC dropout需要将模型转发$N$次,这导致$N$慢$N$次。对于自动驾驶汽车系统等实时应用,需要尽可能快地获得预测和不确定性,因此MC丢弃变得不切实际。在这项工作中,我们提出了基于区域的时间聚合(RTA)的方法,利用视频中的时间信息来模拟采样过程。我们使用提拉米苏主干的RTA方法比使用提拉米苏主干的MC丢弃($N=5$)快10倍。此外,通过我们的RTA方法获得的不确定性估计与MC dropout在像素级和帧级度量上的不确定性估计相当。

英文摘要:

Uncertainty estimation in deep learning becomes more important recently. A deep learning model can't be applied in real applications if we don't know whether the model is certain about the decision or not. Some literature proposes the Bayesian neural network which can estimate the uncertainty by Monte Carlo Dropout (MC dropout). However, MC dropout needs to forward the model $N$ times which results in $N$ times slower. For real-time applications such as a self-driving car system, which needs to obtain the prediction and the uncertainty as fast as possible, so that MC dropout becomes impractical. In this work, we propose the region-based temporal aggregation (RTA) method which leverages the temporal information in videos to simulate the sampling procedure. Our RTA method with Tiramisu backbone is 10x faster than the MC dropout with Tiramisu backbone ($N=5$). Furthermore, the uncertainty estimation obtained by our RTA method is comparable to MC dropout's uncertainty estimation on pixel-level and frame-level metrics.

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。

↑