Abstract: For single-view clothed human reconstruction, the fashionable PIFu-like framework depends on the pixel-aligned feature essentially, while this leads to depth ambiguity and inaccuracy of ...