Get Started
Home
Topics
Search
Library
Research questionHow can sparse SDF observations constrain a frozen INR’s latent code to complete plausible unseen 3D geometry?Sparse off-grid SDF samples can fit a decoder while leaving its per-instance latent code underconstrained. Different latent codes may agree with observed samples yet imply implausible geometry elsewhere.
AI
Computer Vision
Machine Learning
Neural and Evolutionary Computing
Research Paper
Technology
Latest papersRecent research connected to this question, newest first.Observation-Conditioned Latent Energy Priors for Sparse Implicit Neural Shape CompletionThe source studies autodecoder-style implicit neural representations with shared coordinate decoders and per-instance latent codes. Evidence comes from a controlled cell-nucleus SDF dataset and a public MedShapeNet-derived SDF completion dataset, comparing post-hoc latent-prior objectives without retraining the frozen decoder.research paper · Sep 3, 2026
Related questions
How can we complete unseen 3D scene regions from sparse, unconstrained views without dense 3D supervision?How can sparse structure-from-motion remain reliable in repeated or symmetric scenes while limiting redundant computation?How can neural fields reconstruct missing signal regions from sparse observations while retaining reusable task priors?How can sparse-view 3D scene representations avoid cross-view artifacts without costly per-scene reconstruction?