magic factors when upsample flow

[ahangchen]

There are some magic factors when upsample flow to higher resolution： https://github.com/ltkong218/FastFlowNet/blob/main/models/FastFlowNet.py#L140

https://github.com/ltkong218/FastFlowNet/blob/main/models/FastFlowNet.py#L147

https://github.com/ltkong218/FastFlowNet/blob/main/models/FastFlowNet.py#L154

https://github.com/ltkong218/FastFlowNet/blob/main/models/FastFlowNet.py#L161

What's the meaninig of 0.625, 1.25, 2.5, 5? Is there any geometry motivation?

I think the factors should be 2, because when you upsample a flow to a resolution with double height and width, the flow is double due to double pixels between origin points and corresponding points.

[ltkong218]

Please note that each decoder of FastFlowNet estimates optical flow whose magnitude is 1/20 of ground truth flow, therefore, to convert to true displacement before warping, we should multiply a scale factor of 20/(2^l), where l belongs to {2, 3, 4, 5} means the pyramid level. We adopt the same scaling approach as PWC-Net and LiteFlowNet.