你即將離開本站

並前往https://openreview.net/pdf?id=DF8LCjR03tX

HRFormer: High-Resolution Transformer for Dense Prediction

HRFormer-B + OCR gains +1.2% and. +2.0% mIoU over HRNet-W48 + OCR [54] with 25% fewer parameters and slightly more FLOPs on. PASCAL-Context test and COCO-Stuff ...

確定！ 回上一頁

查詢 「OCR HRNet」的人也找了：

  1. Lite-HRNet
    2.
  2. HRNet github
    3.
  3. Semantic segmentation
    4.
  4. hierarchical multi-scale attention for semantic segmentation
    5.
  5. Semantic segmentation state of the art
    6.
  6. HR Net
    7.
  7. Semantic segmentation GitHub
    8.
  8. deep high-resolution representation learning for human pose estimation
    9.