FOCUS -- Multi-View Foot Reconstruction From Synthetically Trained Dense Correspondences
This work addresses the problem of efficient and accurate foot reconstruction for applications like healthcare or footwear design, but it is incremental as it builds on existing synthetic datasets and parameterized models.
The paper tackles 3D human foot reconstruction from few multi-view RGB images by introducing a synthetic dataset with dense correspondences and an uncertainty-aware predictor, achieving state-of-the-art quality in few-view settings with comparable performance to existing methods in many-view scenarios and faster runtime.
Surface reconstruction from multiple, calibrated images is a challenging task - often requiring a large number of collected images with significant overlap. We look at the specific case of human foot reconstruction. As with previous successful foot reconstruction work, we seek to extract rich per-pixel geometry cues from multi-view RGB images, and fuse these into a final 3D object. Our method, FOCUS, tackles this problem with 3 main contributions: (i) SynFoot2, an extension of an existing synthetic foot dataset to include a new data type: dense correspondence with the parameterized foot model FIND; (ii) an uncertainty-aware dense correspondence predictor trained on our synthetic dataset; (iii) two methods for reconstructing a 3D surface from dense correspondence predictions: one inspired by Structure-from-Motion, and one optimization-based using the FIND model. We show that our reconstruction achieves state-of-the-art reconstruction quality in a few-view setting, performing comparably to state-of-the-art when many views are available, and runs substantially faster. We release our synthetic dataset to the research community. Code is available at: https://github.com/OllieBoyne/FOCUS