Matchable Image Retrieval by Learning from Surface Reconstruction
This work addresses the gap in retrieval performance for overlapping images in 3D reconstruction, which is crucial for applications like computer vision and robotics, though it is incremental by building on existing CNN and BoW approaches.
The paper tackles the problem of matchable image retrieval for overlapping images in 3D reconstruction by proposing a CNN-based method that uses surface reconstruction data for training, resulting in significant acceleration and outperforming state-of-the-art methods.
Convolutional Neural Networks (CNNs) have achieved superior performance on object image retrieval, while Bag-of-Words (BoW) models with handcrafted local features still dominate the retrieval of overlapping images in 3D reconstruction. In this paper, we narrow down this gap by presenting an efficient CNN-based method to retrieve images with overlaps, which we refer to as the matchable image retrieval problem. Different from previous methods that generates training data based on sparse reconstruction, we create a large-scale image database with rich 3D geometrics and exploit information from surface reconstruction to obtain fine-grained training data. We propose a batched triplet-based loss function combined with mesh re-projection to effectively learn the CNN representation. The proposed method significantly accelerates the image retrieval process in 3D reconstruction and outperforms the state-of-the-art CNN-based and BoW methods for matchable image retrieval. The code and data are available at https://github.com/hlzz/mirror.