Computer vision semantic segmentation generated by video taken from San Francisco Golden Bridge area, trained from Cityscapes dataset the fourth time, which shows improvement over the 3rd round of training. The improvements are in road and pedestrian classifications.
Keywords: computer vision, semantic segmentation, cityscapes dataset, object detection, tensorflow, fully convolutional networks, self-driving car, and autonomous vehicle
See my GitHub for more details:
https://github.com/blei139/Cityscapes...