Pixel2Mesh++: Multi-View 3D Mesh Generation via Deformation
This addresses the problem of generating high-quality 3D meshes from limited images for applications like computer vision and graphics, though it builds incrementally on prior deformation-based approaches.
The paper tackles 3D mesh generation from a few color images by predicting iterative deformations to improve a coarse shape, leveraging cross-view information with a graph convolutional network. Experiments show the model produces accurate 3D shapes well-aligned to arbitrary viewpoints and generalizes across categories, input numbers, and mesh initializations.
We study the problem of shape generation in 3D mesh representation from a few color images with known camera poses. While many previous works learn to hallucinate the shape directly from priors, we resort to further improving the shape quality by leveraging cross-view information with a graph convolutional network. Instead of building a direct mapping function from images to 3D shape, our model learns to predict series of deformations to improve a coarse shape iteratively. Inspired by traditional multiple view geometry methods, our network samples nearby area around the initial mesh's vertex locations and reasons an optimal deformation using perceptual feature statistics built from multiple input images. Extensive experiments show that our model produces accurate 3D shape that are not only visually plausible from the input perspectives, but also well aligned to arbitrary viewpoints. With the help of physically driven architecture, our model also exhibits generalization capability across different semantic categories, number of input images, and quality of mesh initialization.