Uncertainty Estimates and Multi-Hypotheses Networks for Optical Flow
This work addresses uncertainty estimation in optical flow for computer vision applications, offering a novel method that improves reliability without sacrificing speed.
The paper tackles optical flow estimation by introducing networks that estimate their own prediction uncertainty and provide multiple hypotheses, achieving uncertainty quality above previous confidence measures and enabling interactive frame rates.
Optical flow estimation can be formulated as an end-to-end supervised learning problem, which yields estimates with a superior accuracy-runtime tradeoff compared to alternative methodology. In this paper, we make such networks estimate their local uncertainty about the correctness of their prediction, which is vital information when building decisions on top of the estimations. For the first time we compare several strategies and techniques to estimate uncertainty in a large-scale computer vision task like optical flow estimation. Moreover, we introduce a new network architecture utilizing the Winner-Takes-All loss and show that this can provide complementary hypotheses and uncertainty estimates efficiently with a single forward pass and without the need for sampling or ensembles. Finally, we demonstrate the quality of the different uncertainty estimates, which is clearly above previous confidence measures on optical flow and allows for interactive frame rates.