I am building an Android App with uses the DNN module.
The output of the forward pass is a Mat with dims() = 4.
Actually its the output of semantic segmentation (ENet Example) with 1x8x45x80.
mat.total() = 28800 which is also correct.
The problem is that i cannot access the data of this mat with get(0,0,floatbuffer), as cols=-1 and rows=-1 which is normal for dim>2.
Is there any other way? In the end i would like to have 8 Mats (or 8 simple float-arrays) for each plane.
Best Regards
↧
