Abstract

We propose a learning-based approach for motion boundary detection. Precise localization of motion boundaries is essential for the success of optical flflow estimation, as motion boundaries correspond to discontinuities of the optical flflow fifield. The proposed approach allows to predict motion boundaries, using a structured random forest trained on the ground-truth of the MPI-Sintel dataset. The random forest leverages several cues at the patch level, namely appearance (RGB color) and motion cues (optical flflow estimated by state-of-the-art algorithms). Experimental results show that the proposed approach is both robust and computationally effificient. It signifificantly outperforms state-of-the-art motion-difference approaches on the MPISintel and Middlebury datasets. We compare the results obtained with several state-of-the-art optical flflow approaches and study the impact of the different cues used in the random forest. Furthermore, we introduce a new dataset, the YouTube Motion Boundaries dataset (YMB), that comprises 60 sequences taken from real-world videos with manually annotated motion boundaries. On this dataset, our approach, although trained on MPI-Sintel, also outperforms by a large margin state-of-the-art optical flflow algorithms