wayveai / fiery

PyTorch code for the paper "FIERY: Future Instance Segmentation in Bird's-Eye view from Surround Monocular Cameras"
https://wayve.ai/blog/fiery-future-instance-prediction-birds-eye-view
MIT License
561 stars 85 forks source link

Is future_egopose necessary for inference? #36

Closed synsin0 closed 2 years ago

synsin0 commented 2 years ago

Thanks for your great work. I have a little question about future ego pose during inference? I may find a little tricky because flow prediction is a module before motion planning. In real cases, the flow prediction module has no chance of getting future ego pose. But the code may show future ego pose is irreplacable in inference. When I turn to None, the inference doesn't work.

synsin0 commented 2 years ago

I have understood the receptive field in the model, then info from future has been masked. No problems.

viv365 commented 1 year ago

Hy synsin0, can u explain in detail about this future_ego_motion ? i am also having lot of confusion about this !

anthonyhu commented 1 year ago

future_egomotion is used to warp the past features to the present's reference frame, see https://github.com/wayveai/fiery/blob/master/fiery/models/fiery.py#L143