SPengLiang

Results 24 comments of SPengLiang

scipy.optimize.minimize can help you with minimizing the reprojection error of keypoints.

#================================= Here are some of my code in my previous works, and it works well. You can customize it to fit your code, such as corner order and optimizer. #=================================...

@maudzung You're quite welcome. You probably need to change Ry to -Ry as the orientation I used may be different from KITTI's, and the order of keypoints in the optimize...

@atakagi-fixstars thank you,I tried many ways but still failed,My CUDA version is 9.0 and Visual Studio version is VS2015

@atakagi-fixstars First thank you,I used the listed command in windows command line,because my OS is windows ,so I repalce the command “make” by using windows minGW。

It seems that the symlink has been created, you can delete these symlinks or this symlink directory and try again.

Thanks for your interest in this work. 1. Yes, the pre-trained model is evaluated on the KITTI validation set. 2. According to our results, our work has a similar AP...

Thanks for your interest in this work. Similar to CaDDN, we also convert Waymo data format to KITTI for convenience in training. Most codes are the same except the dataloader....

We scale the visual depth according to the scale factor, as shown in Figure 3 in the paper.

Thanks for your interest in this work. 1. Yes; 2. You are right. In this codebase, we directly use the generated pseudo labels for convenience, as the gap between pseudo...