Fig. 5From: Structure from motion using dense CNN features with keypoint relocalizationExample of 3D reconstruction in the Aachen dataset. These figures show qualitative examples of SfM using DoG+RootSIFT [6] (a) and our dense CNN with keypoint relocalization (b). Our method can reconstruct all the 21 images in the subset whereas the baseline DoG+RootSIFT [6] fails to reconstruct it. As a nature of dense feature matching, our method reconstructs 42,402 3D points which are 8.2 times more than the baseline methodBack to article page