AlexeyAB / yolo2_light

Light version of convolutional neural network Yolo v3 & v2 for objects detection with a minimum of dependencies (INT8-inference, BIT1-XNOR-inference)

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

detect result is not right

fidodido2503 opened this issue · comments

When I increase the image size ,the detection result is not right.
And in Windows 10 result is right, But in Ubuntu 16.04, program get wrong result.
I tested image size(1920,1088), and the GPU version.

I find this line went wrong
forward_maxpool_layer_kernel << <cuda_gridsize(n), BLOCK >> > (n, layer.h, layer.w, layer.c, layer.stride, layer.size, layer.pad, state.input, layer.output_gpu, layer.indexes_gpu);

@fidodido2503 What is wrong in this line?