Convert to NCNN

[Siziff]

How to convert the model to ncnn? If there is no WARP layer in ncnn, then how can it be converted with a dimension of not 3 channels? in the inference function, the flow dimension has 2 channels instead of the required three to replace, as they do here - https://github.com/TNTwise/REAL-Video-Enhancer/wiki/Convert-Rife-Models

[ltkong218]

You can refer to ncnn Implementation of IFRNet. It may be helpful.

[Siziff]

@ltkong218 I have seen this solution, but it does not describe the conversion process, but only gives the already converted models :(