New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

[WIP] flownet2-docker comparison #160

Open

BlGene wants to merge 1 commit into NVIDIA:master from BlGene:master

BlGene commented Jun 24, 2019 •

edited

Loading

I've tried to compare the flownet2-pytorch results to flownet2-docker results

for this I created copy/simlink of the flownet2-docker directory in the flownet2-pytorch directory.
In this directory I then ran the flownet2-docker command ./run-network.sh -n FlowNet2 -v data/0000000-imgL.png data/0000001-imgL.png flow.flo
I then ran the new run_pair.py command.

I get a very large difference(L2 norm of 81202.305) between the predicted flows, basically flownet2-pytorch is not working for some reason.

Things that I haven't looked at yet are pre-processing:

RGB vs BGR
Normalization (divide by 255.0)?
multiply by 20?


          install fixes, flownet2-docker comparison

de0cbe7

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet