Search papers, labs, and topics across Lattice.
This paper introduces Hierarchical Gaussian Fields (HierGF) to address sparse-view 3D reconstruction challenges by generating self-supervision from limited observations. A two-stage geometry-perception backbone transforms coarse 3D geometry and 2D generative priors into structured pseudo-supervision, enhancing multi-view consistency. A learnable confidence network and a geometrically consistent densification module further improve reconstruction quality in under-sampled regions.
Sparse-view 3D reconstruction now generates its own reliable supervision from limited geometry and 2D priors, overcoming the limitations of traditional reprojection errors and heuristic density control.
Sparse view 3D reconstruction is an important and common scenario in multimedia applications, such as augmented reality/virtual reality (AR/VR) content creation, cultural heritage digitization, and certain robotic applications, where only a limited number of randomly captured views may be available. However, sparse views contain only limited 3D information, posing two major challenges:1) too few images are available for matching, making it difficult to build multi-view consistency; 2) insufficient view coverage leads to a lack of information in under-sampled regions, resulting in missing parts of object structure. Existing methods mostly still rely on limited reprojection errors and regularization terms, which are prone to overfitting to a single view and inconsistent appearances across views. In geometrically under-sampled regions, they often rely on heuristic density control, lacking reliable guidance and often resulting in blurring and structural holes.To address these issues, this paper proposes Hierarchical Gaussian Fields (HierGF), which revisits sparse-view reconstruction from a hierarchical geometry-perception perspective and converts limited observations into reliable self-generated supervision beyond fixed priors and heuristic density control. In particular, we transform coarse 3D geometric information and additional 2D generative priors into structured pseudo-supervision through a two-stage geometry-perception backbone network, thereby enhancing multi-view consistency with very few input views. In addition, we introduce a learnable confidence network to guide gradients toward cross-view consistent content, and a geometrically consistent densification module to improve the reconstruction of multi-view alignment and under-sampled regions.
