Zhengyang Feng

@cardwing Found it (finally) [here](https://github.com/cardwing/Codes-for-Lane-Detection/issues/11#issuecomment-654170702). It was a non-related old and closed issue, I think we should move the discussions here since it is in this paper ERFNet-SAD is introduced....

@stiglioglu We don't have camera APIs, if that is what you want. I think your use case is similar to a video input? We have a video-inference-visualize pipeline: https://github.com/voldemortX/pytorch-auto-drive/blob/master/docs/VISUALIZATION.md#segmentation-mask-video

The general method is read the image, make the testing transforms (e.g., resize), then inference, finally apply specific post-processings. For some lane detection methods, they already have a `inference()` method,...

> where do you read the video frames can you help me i couldn't find it We use a torch style dataloader for video reading: https://github.com/voldemortX/pytorch-auto-drive/blob/master/utils/datasets/video.py

@ccblublu If your custom dataset is already labeled by sample points, you can refer to this documentation: https://github.com/voldemortX/pytorch-auto-drive/blob/master/docs/CURVE.md But you will need to get the [loader](https://github.com/voldemortX/pytorch-auto-drive/blob/7280e135d79b0d3f9c9c57a73293da6bcd3a2bab/tools/curve_fitting_tools/loader.py) working for your dataset....

@Zhuanglong2 Hi! Did you launch a training? can you provide the config file for your model, and some info about your local machine? e.g., CPU, GPU, memory

Also, I think your error message includes something about a numpy error? It could be related. Maybe this can help: https://stackoverflow.com/a/54708388/15449902

> When i run "lane_video.py", i meet this problem"RuntimeError: input must be contiguous". Then, I debug this code, and i find the problem appears in "flipped = self.proj2_conv(flipped, feature)". This...

@Zhuanglong2 Could you try changing this line: https://github.com/voldemortX/pytorch-auto-drive/blob/2b0d5ec5f7536c9d2d2b6d8498718a8fca2ab276/utils/datasets/video.py#L28 to: ``` images = images[..., [2, 1, 0]].permute(0, 3, 1, 2).contiguous() / 255.0 # BHWC-rgb uint8 -> BCHW-rgb float ``` In my...

Thanks for your interest in our work! Happy Mid-Autumn Festival!