Rating is available when the video has been rented.
This feature is not available right now. Please try again later.
Published on Jan 18, 2018
The video shows RGB camera images which are pixel-wisely segmented by a CNN (FCN8s). The FCN8s was trained on the Cityscapes Dataset. The segmented output is further converted into a binary images, labeling static and potential dynamic objects. Only pixels labeled as static are retained for the continuous stereo camera calibration algorithm.