PIFuHD
We introduce a multi-level framework that infers 3D geometry of clothed humans at an unprecedentedly high 1k image resolution in a pixel-aligned manner, retaining the details in the original inputs without any post-processing. Recent advances in image-based 3D human shape estimation have been driven by the significant improvement in representation power afforded by deep neural networks. Although c
PIFu
Our approach can digitize intricate variations in clothing, such as wrinkled skirts, high-heels, and complex hairstyles. Shape and textures can be fully recovered in largely unseen regions such as the back of the subject. Our method can also be extended to multi-view input images. We introduce Pixel-aligned Implicit Function (PIFu), a highly effective implicit representation that locally aligns pi
1
リリース、障害情報などのサービスのお知らせ
最新の人気エントリーの配信
処理を実行中です
j次のブックマーク
k前のブックマーク
lあとで読む
eコメント一覧を開く
oページを開く
