# SuperPrimitive: Scene Reconstruction at a Primitive Level

arXiv: https://arxiv.org/abs/2312.05889 · HTML: https://mlsanity.com/paper/2312.05889/

## Key facts

- Cited by 11 papers in the map (data through 14 Oct 2026).
- Builds on 34 of its 58 references that are in the map.
- First posted on arXiv 10 Dec 2023.

## Builds on (most cited first, 30 of 34)

- [PyTorch: An Imperative Style, High-Performance Deep Learning Library](https://mlsanity.com/paper/1912.01703/) (2019)
- [Segment Anything](https://mlsanity.com/paper/2304.02643/) (2023)
- [Vision Transformers for Dense Prediction](https://mlsanity.com/paper/2103.13413/) (2021)
- [Towards Robust Monocular Depth Estimation: Mixing Datasets for Zero-shot Cross-dataset Transfer](https://mlsanity.com/paper/1907.01341/) (2019)
- [Depth Map Prediction from a Single Image using a Multi-Scale Deep Network](https://mlsanity.com/paper/1406.2283/) (2014)
- [ORB-SLAM3: An Accurate Open-Source Library for Visual, Visual-Inertial and Multi-Map SLAM](https://mlsanity.com/paper/2007.11898/) (2020)
- [RegNeRF: Regularizing Neural Radiance Fields for View Synthesis from Sparse Inputs](https://mlsanity.com/paper/2112.00724/) (2021)
- [Hypersim: A Photorealistic Synthetic Dataset for Holistic Indoor Scene Understanding](https://mlsanity.com/paper/2011.02523/) (2020)
- [Ref-NeRF: Structured View-Dependent Appearance for Neural Radiance Fields](https://mlsanity.com/paper/2112.03907/) (2021)
- [Direct Sparse Odometry](https://mlsanity.com/paper/1607.02565/) (2016)
- [NeuralRecon: Real-Time Coherent 3D Reconstruction from Monocular Video](https://mlsanity.com/paper/2104.00681/) (2021)
- [Non-Local Spatial Propagation Network for Depth Completion](https://mlsanity.com/paper/2007.10042/) (2020)
- [DeepV2D: Video to Depth with Differentiable Structure from Motion](https://mlsanity.com/paper/1812.04605/) (2018)
- [CodeSLAM - Learning a Compact, Optimisable Representation for Dense Visual SLAM](https://mlsanity.com/paper/1804.00874/) (2018)
- [DeepFactors: Real-Time Probabilistic Dense Monocular SLAM](https://mlsanity.com/paper/2001.05049/) (2020)
- [Estimating and Exploiting the Aleatoric Uncertainty in Surface Normal Estimation](https://mlsanity.com/paper/2109.09881/) (2021)
- [TartanVO: A Generalizable Learning-based VO](https://mlsanity.com/paper/2011.00359/) (2020)
- [TransformerFusion: Monocular RGB Scene Reconstruction using Transformers](https://mlsanity.com/paper/2107.02191/) (2021)
- [PlaneRCNN: 3D Plane Detection and Reconstruction from a Single Image](https://mlsanity.com/paper/1812.04072/) (2018)
- [Fusion++: Volumetric Object-Level SLAM](https://mlsanity.com/paper/1808.08378/) (2018)
- [Unsupervised Depth Completion from Visual Inertial Odometry](https://mlsanity.com/paper/1905.08616/) (2019)
- [QuadricSLAM: Dual Quadrics from Object Detections as Landmarks in Object-oriented SLAM](https://mlsanity.com/paper/1804.04011/) (2018)
- [SimpleRecon: 3D Reconstruction Without 3D Convolutions](https://mlsanity.com/paper/2208.14743/) (2022)
- [How do neural networks see depth in single images?](https://mlsanity.com/paper/1905.07005/) (2019)
- [PlaneNet: Piece-wise Planar Reconstruction from a Single RGB Image](https://mlsanity.com/paper/1804.06278/) (2018)
- [NodeSLAM: Neural Object Descriptors for Multi-View Shape Reconstruction](https://mlsanity.com/paper/2004.04485/) (2020)
- [Unsupervised Depth Completion with Calibrated Backprojection Layers](https://mlsanity.com/paper/2108.10531/) (2021)
- [Learning Topology from Synthetic Data for Unsupervised Depth Completion](https://mlsanity.com/paper/2106.02994/) (2021)
- [Monitored Distillation for Positive Congruent Depth Completion](https://mlsanity.com/paper/2203.16034/) (2022)
- [Monocular Visual-Inertial Depth Estimation](https://mlsanity.com/paper/2303.12134/) (2023)

## Built on by (most cited first, 11 of 11)

- [MASt3R-SLAM: Real-Time Dense SLAM with 3D Reconstruction Priors](https://mlsanity.com/paper/2412.12392/) (2024)
- [ViSTA-SLAM: Visual SLAM with Symmetric Two-view Association](https://mlsanity.com/paper/2509.01584/) (2025)
- [G4Splat: Geometry-Guided Gaussian Splatting with Generative Prior](https://mlsanity.com/paper/2510.12099/) (2025)
- [SPEAR: A Simulator for Photorealistic Embodied AI Research](https://mlsanity.com/paper/2607.06701/) (2026)
- [FusionSense: Bridging Common Sense, Vision, and Touch for Robust Sparse-View Reconstruction](https://mlsanity.com/paper/2410.08282/) (2024)
- [MultiDepth: Multi-Sample Priors for Refining Monocular Metric Depth Estimations in Indoor Scenes](https://mlsanity.com/paper/2411.01048/) (2024)
- [4D Primitive-Mach'e: Glueing Primitives for Persistent 4D Scene Reconstruction](https://mlsanity.com/paper/2512.16564/) (2025)
- [PlanaReLoc: Camera Relocalization in 3D Planar Primitives via Region-Based Structure Matching](https://mlsanity.com/paper/2603.20818/) (2026)
- [DecoRec: Decomposed 3D Scene Reconstruction from Single-View Images via Object-Level Diffusion](https://mlsanity.com/paper/2605.16807/) (2026)
- [PixVOD: Pixel-Distributed Direct Visual Odometry and Depth Estimation](https://mlsanity.com/paper/2606.03989/) (2026)
- [BLASt3R: Bundle Adjustment of Any Image Set with Multi-View Matching and Monocular Priors](https://mlsanity.com/paper/2609.05210/) (2026)

## Abstract (arXiv)

Joint camera pose and dense geometry estimation from a set of images or a monocular video remains a challenging problem due to its computational complexity and inherent visual ambiguities. Most dense incremental reconstruction systems operate directly on image pixels and solve for their 3D positions using multi-view geometry cues. Such pixel-level approaches suffer from ambiguities or violations of multi-view consistency (e.g. caused by textureless or specular surfaces). We address this issue with a new image representation which we call a SuperPrimitive. SuperPrimitives are obtained by splitting images into semantically correlated local regions and enhancing them with estimated surface normal directions, both of which are predicted by state-of-the-art single image neural networks. This provides a local geometry estimate per SuperPrimitive, while their relative positions are adjusted based on multi-view observations. We demonstrate the versatility of our new representation by addressing three 3D reconstruction tasks: depth completion, few-view structure from motion, and monocular dense visual odometry.