Open Jian-danai opened 11 months ago
https://github.com/DerryHub/BEVFormer_tensorrt/blob/303d3140c14016047c07f9db73312af364f0dd7c/det2trt/models/modules/encoder.py#L256C10-L258C84
bev_mask = (1 - (1 - bev_mask).prod(0)).view(6, -1, 1) bev_mask = bev_mask / torch.clamp(bev_mask.sum(0, keepdims=True), min=1e-4)
What's the purpose of this, it makes bev_mask different from original point_sampling function.
bev_mask
point_sampling
Where did you handle those difference?[It seems bev_mask will not be used later, so it does not matter?]
Another question, https://github.com/DerryHub/BEVFormer_tensorrt/blob/303d3140c14016047c07f9db73312af364f0dd7c/det2trt/models/modules/encoder.py#L297
shift_ref_2d = ref_2d.clone() shift_ref_2d = shift_ref_2d + shift.view(1, 1, 1, 2) * use_prev_bev
If use clone() there, it will not match original BEVformer, right? Original ref_2d will change when shift_ref_2d changing. Besides, Original shift_ref_2d always add shift.view(1, 1, 1, 2) even if in the first sample(i.e. when use_prev_bev=0).
clone()
ref_2d
shift_ref_2d
shift.view(1, 1, 1, 2)
use_prev_bev=0
https://github.com/DerryHub/BEVFormer_tensorrt/blob/303d3140c14016047c07f9db73312af364f0dd7c/det2trt/models/modules/encoder.py#L256C10-L258C84
What's the purpose of this, it makes
bev_maskdifferent from originalpoint_samplingfunction.Where did you handle those difference?[It seems
bev_maskwill not be used later, so it does not matter?]Another question, https://github.com/DerryHub/BEVFormer_tensorrt/blob/303d3140c14016047c07f9db73312af364f0dd7c/det2trt/models/modules/encoder.py#L297
If use
clone()there, it will not match original BEVformer, right? Originalref_2dwill change whenshift_ref_2dchanging. Besides, Originalshift_ref_2dalways addshift.view(1, 1, 1, 2)even if in the first sample(i.e. whenuse_prev_bev=0).