← 返回卡片

原书出处

Foundations of Computer Vision (Antonio Torralba, Phillip Isola etc.)

核验版本的摘录快照 · 导入于 2026-11-14T15:40:56.582Z

Johansson [238] demonstrated that by watching the motion of only a few of those points, no more than 12, it was possible to clearly recognize the person actions. By showing only moving points, Johansson dissociated the perception of form/shape from the motion pattern. Johansson's videos were very surprising at the time and researchers were stunned by how easy was to recognize many types of actions

此预览只包含定位所需的摘录和上下文。

原书定位详情

来源:a19c7933-4eda-4176-aebd-30fe7e1ad292

正文位置:extracted/Foundations of Computer Vision (Antonio Torralba, Phillip Isola etc.)/_epub-src/OPS/xhtml/c044.xhtml

EPUB SHA-256:60a7eebc43f5d9a675d776fed3736fbdaa791e55877023f2f8cba016a9b384f8