Seokju Lee

[VPGNet](https://arxiv.org/abs/1710.06288) dataset is available. It contains around 21K images with human-annotated lane, road markings, and vanishing point labels. For more details, please check our project [page](https://github.com/SeokjuLee/VPGNet#vpgnet-dataset).

Can I make a layer of which input is disparity & output is Rt?

3

With a given disparity map between i1 and i2, I want to generate the relative pose. Can I make this layer with GVNN library? If it is possible, could you...

About the pose from SensorData.py

3

I am preprocessing the relative pose between color frames (t, t+1). Is the pose coordinate aligned to the camera coordinate, defined as CameraToWorld? If then, is it right to generate...

[VPGNet](https://arxiv.org/abs/1710.06288) dataset is available. It contains around 21K images with human-annotated lane, road markings, and vanishing point labels. For more details, please check our project [page](https://github.com/SeokjuLee/VPGNet#vpgnet-dataset).

Notation of depth map

1

Hi, I have a question about the notation of depth maps (**computed_depth**, **projected_depth**) in 'inverse_warp2' function in 'inverse_warp.py'. Suppose that D_a is **ref_depth** and D_b is **tgt_depth** as in your...

Dataset

14

VPGNet dataset is available. Please fill out a [form](http://forms.gle/LNCPUgEu4B7XGjLZA) for the download link, and check the README. Thanks.

About post-processing for lane visualization

45

Here is an additional explanation about the lane post-processing. There are four steps basically. First, sampling seed points with the lane heat map from the multi-label task. Then, cluster the...

Code issues

55

Please ask installation, training and test issues in this panel.