alexmonti19 / dagnet Goto Github PK

Hello! Thank you for your poject!

I downloaded npy data from
When I opened the npy data, it's shape was (14032, 50, 11, 4).
Can I know what each number means? According to your paper, I think that 14032 is the number of sequences and 50 is the length of each sequence, but I don't know what 11 and 4 means.

Thank you!

We found that the effect you achieved on the SDD data set is significantly better than other methods, but why are the results of the same SDD data set so different?


Can the results of the two be compared uniformly? Hope you can clear up my current confusion.

I wish you have a nice day！

Hi,
I have just learned something about Motion Trajectory Prediction, i feel confused that can this work support first-person perspective? I find lots of papers show their results just from god perspective, can these papers work well when i use them from first-person perspective?

How is the --run parameter used? I wish to execute evaluate_sdd.py but /runs/dagnet is empty.
I tried with exp_sdd_best from pretrained models still gettibg this error
FileNotFoundError: [Errno 2] No such file or directory: '/home/runs/dagnet/exp_sdd_best'

@alexmonti19
Thanks for your hard work!

Take this example, i want to predict the order of the dialog in a comic book, basically telling which dialog box is 1st, which is 2nd, 3rd, etc...

• Can your code be trained to solve such task. wont the dialog boxes all look the same to it
• Can it be trained to predict 40+ dialog boxes in a single page. and at what computation cost.
• Can it generalize on unseen images. Will it really understand the visual context since most dialog boxes look the same.

Note that i already have the dialog boxes location and coordinated detected, but now i only want to predict their reading-order.