Sparse-View 3D Reconstruction: Recent Advances and Open Challenges
This is an incremental survey that provides a unified perspective on methods for sparse-view 3D reconstruction, which is essential for applications like robotics and AR/VR where dense image acquisition is impractical.
This survey reviews recent advances in sparse-view 3D reconstruction methods, analyzing how neural implicit models, explicit point-cloud approaches, and hybrid frameworks address challenges like minimal image overlap and artifacts, with comparative results on standard benchmarks revealing trade-offs between accuracy, efficiency, and generalization.
Sparse-view 3D reconstruction is essential for applications in which dense image acquisition is impractical, such as robotics, augmented/virtual reality (AR/VR), and autonomous systems. In these settings, minimal image overlap prevents reliable correspondence matching, causing traditional methods, such as structure-from-motion (SfM) and multiview stereo (MVS), to fail. This survey reviews the latest advances in neural implicit models (e.g., NeRF and its regularized versions), explicit point-cloud-based approaches (e.g., 3D Gaussian Splatting), and hybrid frameworks that leverage priors from diffusion and vision foundation models (VFMs).We analyze how geometric regularization, explicit shape modeling, and generative inference are used to mitigate artifacts such as floaters and pose ambiguities in sparse-view settings. Comparative results on standard benchmarks reveal key trade-offs between the reconstruction accuracy, efficiency, and generalization. Unlike previous reviews, our survey provides a unified perspective on geometry-based, neural implicit, and generative (diffusion-based) methods. We highlight the persistent challenges in domain generalization and pose-free reconstruction and outline future directions for developing 3D-native generative priors and achieving real-time, unconstrained sparse-view reconstruction.