Yvette Zhao

@FrancisGee How's your progress now?

@XingangPan Hi Xinggang, from your paper, you said the model is based on DeepLab-LargeFOV VGG version. But from your github main page, you said your code is modified from ResNet,...

@XingangPan ok, thank you for your reply. I figured out it's based on VGG. I saw the printout of the VGG pretrained model from your link [here](https://drive.google.com/file/d/12RLXY6o8gaGMY1K1g6d447Iby9ewVIyV/view). It's modified with...

@ycszen sorry, nope, I switch to Caffe now.

@cardwing That's great, thanks!

@cuicuizhang1989 would like to know FrancisGee's solution works for you or not?

@Neha-swan find the solution [here, issue #23 ](https://github.com/XingangPan/SCNN/issues/23). BTW, pay attention to that, your opencv should be **Opencv (for tools/lane_evaluation), version 2.4.8 (later 2.4.x should also work).**.

@XingangPan Hi Xingang, I embedded the SCNN module inside VGG and train this model with CULane data on V100(4*16G). Even though I set the batch size 8, there's still an...

@LokLu Hi, I met the same problem, do you figure why it does so? Thanks :)

百度的 Apolloscapes 中 lane_segmentation 任务里面有路面的交通标识：http://apolloscape.auto/lane_segmentation.html