Bz、
Bz、
Hi, Thank you for your good work! Recently i am studying DETR-like, and i have questions as following: 1.there is no implement about reference point iterative refinement , but many...
I have a question, does the model can support multi-task? Or the architecture can support multi-task, but you can only choose one specific task when you train.
i find the batch fix to 1, what is the reason?
I find the comannd is generated by ego future position,but I can't find where is the high-command used in model。I find cmd is encoded as an embedding and concated with...