Microsoft's MoGe-3 Sets New SOTA on Depth Estimation Benchmarks
NielsRogge · x · 2026-08-31
Microsoft released MoGe-3, the new state-of-the-art on multiple depth estimation benchmarks on Papers with Code, in the paper "MoGe-3: Fine-Detail Monocular Geometry Estimation with Self-Guided Sparse Volumetric Refinement".
Key idea: current SOTA models decode 3D geometry in a 2D parameterization where feature interactions are governed by image-plane proximity rather than true 3D spatial relations, mixing features from geometrically distant surfaces and over-smoothing thin structures. MoGe-3 introduces Self-Guided Sparse 3D Refinement (SSR): it lifts the coarse point map from a foundation base model onto a sparse voxel shell and refines it with sparse convolutions that aggregate features by 3D spatial locality, avoiding feature mixing across depth discontinuities to produce high-fidelity, metric-scale point maps.
Experiments across diverse datasets show it significantly outperforms existing approaches in recovering fine 3D detail, both quantitatively and visually. Weights are open on Hugging Face (microsoft/moge), with model cards in progress.
More from Research
- Interpretability debate: one interpretable polysemantic neuron doesn't prove all are — aran_nayebi · 2026-10-16
- Optuna creator Akiba: when 2 hyperparameters cost $20M, what's left for BO/HPO — Stefania_druga · 2026-10-16
- AI will be superhuman at science regardless, and Europe can't opt out — teortaxesTex · 2026-10-16
- SERO augments long-horizon agentic RL with executable oracle continuations — Shanghai-AI-Laboratory · 2026-10-16
- Tsinghua and Amap propose OmniVerifier-C1, a graded verifier beating binary judges — Bowei Liu · 2026-10-16
- Alibaba open-sources MAI-Gym, an all-in-one GUI agent eval and RL training platform — Tongyi-MAI · 2026-10-16