2P1-E05 投票によるSIFTを用いた物体移動量計測

概要

論文の詳細を見る
This paper presents an algorithm that derives rigid body location using SIFT (Scale-Invariant Feature Transform) and voting technique. This algorithm derives rigid body location corresponding to feature points between two images. First, SIFT detects corresponding feature points between two images. Next, the algorithm chooses two pairs of corresponding points at random between two images. The included angle and scale are derived from these two pairs. This computation is iterated to decide the value of angle and scale by voting. The rigid body translation is derived by using a pair of corresponding points and decided angle and scale. This computation is iterated to decide the value of translation by voting. We verify that this algorithm derives correctly rigid body location using artificial images and real images.
一般社団法人日本機械学会の論文