GroupNet
GroupNet copied to clipboard
[CVPR22] GroupNet: Multiscale Hypergraph Neural Networks for Trajectory Prediction with Relational Reasoning
Hi, thank you very much for your contribution to the trajectory prediction community. Your groupNet is wonderful. Regarding the reusability of this code, I wonder when you plan to publish...
hey bro, Do you have plans to release the codes of SDD/ETH-UCY?
1. Changed the `vis_result` function by assigning the previous_3D variable before calling the `draw_result` function ( both of which lie in `test_nba.py` ). 2. Check for `vis/nba/` folder for storing...
Hi, Thanks for the nice work! As there was [a bug in the data processing of Trajectron++](https://github.com/StanfordASL/Trajectron-plus-plus/issues/53), I am wondering whether you have some updated results for GroupNet+Trajectron++ for ETH/UCY...
previous_3D = np.reshape(previous_3D,(batch,actor_num,args.future_length, 2)) draw_result(best_guess,previous_3D) draw_result(gt,previous_3D,mode='gt') return soft_max_1d = F.softmax(trans_input) Traceback (most recent call last): File "**test_nba.py**", line 276, in **vis_result(test_loader, args)** **File "test_nba.py", line 120, in vis_result** **previous_3D =...
Thank you for your wonderful work! Can you share the code for the ETH-UCY and SDD dataset dataloader? Thank you.