Texture Generation Using Dual-Domain Feature Flow with Multi-View Hallucinations
This work addresses the challenge of texture generation for 3D animation rendering from single images, offering a novel method that enhances realism and applicability in computer graphics.
The paper tackles the problem of estimating a complete texture map for 3D human models from a single image, which is insufficient due to limited views, by proposing a dual-domain generative model that generates multi-view hallucinations and texture maps simultaneously, resulting in consistent colors and patterns and improved generation quality for pose and viewpoint transfer tasks.
We propose a dual-domain generative model to estimate a texture map from a single image for colorizing a 3D human model. When estimating a texture map, a single image is insufficient as it reveals only one facet of a 3D object. To provide sufficient information for estimating a complete texture map, the proposed model simultaneously generates multi-view hallucinations in the image domain and an estimated texture map in the texture domain. During the generating process, each domain generator exchanges features to the other by a flow-based local attention mechanism. In this manner, the proposed model can estimate a texture map utilizing abundant multi-view image features from which multiview hallucinations are generated. As a result, the estimated texture map contains consistent colors and patterns over the entire region. Experiments show the superiority of our model for estimating a directly render-able texture map, which is applicable to 3D animation rendering. Furthermore, our model also improves an overall generation quality in the image domain for pose and viewpoint transfer tasks.