chengy12 / disdis Goto Github PK

Hi @CHENGY12

Thank you for your great work. I was able to set up and run code quite quickly!

While reproducing results with your evaluation code using pre-processed dataset pickle file from Trajectron++, I noticed that the results were somewhat different.

I tried to analyze the cause of this, and I found that data_utils.py in your code is based on the eccv2020 branch of trajectron++. In that branch, there was a critical issue in data pre-processing that future two GT coordinates were additionally used while calculating the acceleration of the last observation points. Many issues have been raised in Trajectron++ regarding this problem (issues #26,#40 and #53).

Fortunately, this issue was solved in the Trajectron++ code. Below are my PCMD results after fixing it and re-training the DisDis model:

Note that DisDis still performs much better than other comparable models (including reproduced Trajectron++). Also, I did not edit the config file, so there is room for better results when modifying the hyper-parameters.

As your paper compared the DisDis with other SOTA models which use Social-GAN data-loader (Social-GAN, STGAT, Social-STGCNN), I think the authors should fix this issue and update the numbers in the arXiv paper for a fair comparison.

Thank you.

Are you planning to show the model & training code?

Hi,

Thanks for your good job. I want to ask how to plot the trajectories in scenes in your paper fig5 and how to find which frame corresponds to the scene at this time.?

I am very confused about the visualization about fig5. Can you tell me how to draw this kind of picture? Hope you can help me solve the questions.

Best,
Jincan