cfzd
cfzd
Hi, If you want to infer on a custom dataset, you can try to use these codes: https://github.com/cfzd/Ultra-Fast-Lane-Detection-v2/blob/f666202d39aedcc624248e65dd9c0604b1c6ac8c/evaluation/eval_wrapper.py#L510-L512 In the code above, it will generate a text file containing the...
@HesstisSeca In the first image, it shows that the OpenCV lib is not found. Is the OpenCV correctly installed? Are the bin path, lib path and include path can be...
@Durobert You can either directly use UFLDv1, or you can also both use the row and column anchors as an ensemble learning approach. On the Curvelanes dataset, we use both...
@Durobert I think of course the v2 method should be better.
@578223592 你可以理解为Tusimple和CULane可能相比于Curvelanes更简单一点（它们只有4条线，而Curvelanes最多可能会有10条）。另一个原因是我们希望v2版本和v1在Tusimple和CULane上保持相同的速度。当然你如果在Tusimple和CULane也使用行列一起用的形式的话，性能可能会更好。
@Gannis246 这个是因为我们的预处理里面还包含了一个top_crop的过程，简单来说就是把无用的天空像素给去掉了，你可以参看这部分的代码： https://github.com/cfzd/Ultra-Fast-Lane-Detection-v2/blob/c80276bc2fd67d02579b6eeb57a76cb5a905aa3d/demo.py#L87-L91 https://github.com/cfzd/Ultra-Fast-Lane-Detection-v2/blob/c80276bc2fd67d02579b6eeb57a76cb5a905aa3d/data/dataset.py#L30-L32
@Mollylulu @shuizaola Hi, sorry for the late. I have pushed the code : )
@wjywjy14 have you successfully installed the CUDA extension? ``` cd my_interp sh build.sh # If this fails, you might need to upgrade your GCC to v7.3.0 ```
@wjywjy14 不好意思，我的意思不是CUDA的版本，而是你安装成功我们这个`my_interp`的CUDA拓展了没有，安装方式就是上面写的`sh build.sh`。
@wjywjy14 你可以找一下是报什么错，如果是一些类型转换不对的错的话，可能得需要升级gcc，或者你也可以安装个本地的gcc，临时用一下