The OCR approach is rephrased as Segmentation Transformer: https://arxiv.org/abs/1909.11065. This is an official implementation of semantic segmentation for HRNet. https://arxiv.org/abs/1908.07919
sorry to bother ,but beacause os some reasons i cant run the code by myself ,i just want to know the size of the output of model(seg_hrnet),is it 4 times downsampled or the same size of the input image ,cause i didnt see upsanmpler after hrnet ,but a size in a config is (520(in),520(out)).thanks again!
sorry to bother ,but beacause os some reasons i cant run the code by myself ,i just want to know the size of the output of model(seg_hrnet),is it 4 times downsampled or the same size of the input image ,cause i didnt see upsanmpler after hrnet ,but a size in a config is (520(in),520(out)).thanks again!