cfzd
cfzd
@KaiTang98 In fact, we get the information of (1 1 1 1) from the segmentation map. This label matches the segmentation label exactly. For example, if a segmentation map has...
You are welcome. : )
@awethaileslassie The 200 grids are used for localization. The extra one grid is used for represneting existence of lane points.
@Zengtz I'm not very sure to understand your question. Do you mean that you want to run the model on Ubuntu while the data is on Windows? Why not just...
@Zengtz If you have installed opencv, you can skip this step and directly build the CULane evaluation tool.
@qiyuefun 可以把已训好的模型作为预训练参数载入进去然后在新的数据上进行迁移学习。你可以直接使用fintune部分的逻辑： https://github.com/cfzd/Ultra-Fast-Lane-Detection/blob/85eda710482eebe1a48f90316de7cfec7399bbf3/train.py#L122-L129
@mimi37 Could you please provide more information?
我已经在UFLDv2的仓库中回答了这个问题。 https://github.com/cfzd/Ultra-Fast-Lane-Detection-v2/issues/56
@zqs1996 在数据处理时，我们默认会把车道线做延伸至图像底部的处理，你可以注释掉它，在这个部分： https://github.com/cfzd/Ultra-Fast-Lane-Detection/blob/85eda710482eebe1a48f90316de7cfec7399bbf3/data/dataset.py#L135-L164
@farshidrayhanuiu Please see the new UFLDv2 version. https://github.com/cfzd/Ultra-Fast-Lane-Detection-v2