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
According to code, it seems that multi_scale_output is never false, but within the _make_stage function, the comment indicate that "multi_scale_output is only used last module". Am I missing something?
According to code, it seems that multi_scale_output is never false, but within the _make_stage function, the comment indicate that "multi_scale_output is only used last module". Am I missing something?