agrimgupta92 / sgan

Code for "Social GAN: Socially Acceptable Trajectories with Generative Adversarial Networks", Gupta et al, CVPR 2018
MIT License
819 stars 260 forks source link

How To Show Visualization Result? Thanks. #23

Open ChenYu-Liu opened 5 years ago

ChenYu-Liu commented 5 years ago

HI~ Excuse me, How do i to show visualization result of this model?

trungmanhhuynh commented 5 years ago

Same question, although it is somehow mentioned in comment #5, but I still dont get the homography matrix work properly (from pixels to meters and vice versa).

Anyone please help.

marsmarcin commented 5 years ago

I just make 4 target Visual-of-Social-GAN 4000

trungmanhhuynh commented 5 years ago

@marsmarcin you're the man :)

amiryanj commented 5 years ago

Thank you @marsmarcin for your effort. I made some changes, so that all test samples can be visualized one by one:

plot_model01-new.py.txt

marsmarcin commented 5 years ago

Thank you @marsmarcin for your effort. The above link has problem, then I paste it here again: https://github.com/marsmarcin/Visual-of-Social-GAN I also made some changes, so that all test samples can be visualized one by one:

plot_model01-new.py.txt

hhh,thanks!

trungmanhhuynh commented 5 years ago

I posted the processed data in pixels, meters, and scripts to convert between them on my site: https://github.com/trungmanhhuynh/Scene-LSTM With this, you are able to correctly visualize any trajectories on images.

yeyingdege commented 4 years ago

I posted the processed data in pixels, meters, and scripts to convert between them on my site: https://github.com/trungmanhhuynh/Scene-LSTM With this, you are able to correctly visualize any trajectories on images.

@trungmanhhuynh How do you get the homography matrix of these dataset? Can you explain more about the data process method, since you have many scripts?