Semantic motion segmentation for urban dynamic scene understanding

2016 IEEE International Conference on Automation Science and Engineering (CASE)(2016)

引用 16|浏览58
A mount of recent researches on scene parsing and semantic labeling, while few focus on obtaining joint semantic motion labeling. In this paper, we propose an approach to infer both the object class and motion status for each pixel of images. First, we extract and match sparse image features to estimate ego-motion between two consecutive stereo images, the result of feature points grouping is used to segment moving object in U-disparity map. Second, a Fully Convolutional Neural Network is employed for semantic segmentation. Moreover, semantic cues are utilized to remove pixels have no potential to be moved in motion mask. Finally, we use a fully connected CRF to integrate motion into semantic segmentation. To validate the effectiveness of the proposed algorithm, we present experimental results with KITTI stereo images that contain moving objects.
KITTI stereo images,semantic image segmentation,CRF,fully convolutional neural network,U-disparity map,moving object segmentation,ego-motion estimation,sparse image feature extraction,sparse image feature matching,image pixel,object motion status,object class status,joint semantic motion labeling,semantic labeling,scene parsing,urban dynamic scene understanding,semantic motion segmentation
AI 理解论文