3DMatch: Learning Local Geometric Descriptors from RGB-D Reconstructions. Matching local geometric features on real-world depth images is a challenging task due to the noisy, low-resolution, and incomplete nature of 3D scan data. These difficulties limit the performance of current state-of-art methods, which are typically based on histograms over geometric properties. In this paper, we present 3DMatch, a data-driven model that learns a local volumetric patch descriptor for establishing correspondences between partial 3D data. To amass training data for our model, we propose an unsupervised feature learning method that leverages the millions of correspondence labels found in existing RGB-D reconstructions. Experiments show that our descriptor is not only able to match local geometry in new scenes for reconstruction, but also generalize to different tasks and spatial scales (e.g. instance-level object model alignment for the Amazon Picking Challenge, and mesh surface correspondence). Results show that 3DMatch consistently outperforms other state-of-the-art approaches by a significant margin.
Keywords for this software
References in zbMATH (referenced in 5 articles )
Showing results 1 to 5 of 5.
- de Almeida, Liliane Rodrigues; Giraldi, Gilson Antonio; Vieira, Marcelo Bernardes; Miranda, Gastão Florêncio jun.: Pairwise rigid registration based on Riemannian geometry and Lie structures of orientation tensors (2021)
- Samir, Chafik; Huang, Wen: Coordinate descent optimization for one-to-one correspondence and supervised classification of 3D shapes (2021)
- Chen, Mingjia; Zou, Qianfang; Wang, Changbo; Liu, Ligang: EdgeNet: deep metric learning for 3D shapes (2019)
- Dyke, Roberto M.; Lai, Yu-Kun; Rosin, Paul L.; Tam, Gary K. L.: Non-rigid registration under anisotropic deformations (2019)
- Rezaei, Masoumeh; Rezaeian, Mehdi; Derhami, Vali; Sohel, Ferdous; Bennamoun, Mohammed: Deep learning-based 3D local feature descriptor from mercator projections (2019)