# CODA: Depth-Aligned Scene Completion and Object Decomposition from a Single RGB-D Image

arXiv: https://arxiv.org/abs/2609.25654 · HTML: https://mlsanity.com/paper/2609.25654/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 50 of its 57 references that are in the map.
- First posted on arXiv 22 Sep 2026.

## Builds on (most cited first, 30 of 50)

- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [DINOv2: Learning Robust Visual Features without Supervision](https://mlsanity.com/paper/2304.07193/) (2023)
- [Scalable Diffusion Models with Transformers](https://mlsanity.com/paper/2212.09748/) (2022)
- [Flow Matching for Generative Modeling](https://mlsanity.com/paper/2210.02747/) (2022)
- [Masked-attention Mask Transformer for Universal Image Segmentation](https://mlsanity.com/paper/2112.01527/) (2021)
- [SAM 3: Segment Anything with Concepts](https://mlsanity.com/paper/2511.16719/) (2025)
- [Google Scanned Objects: A High-Quality Dataset of 3D Scanned Household Items](https://mlsanity.com/paper/2204.11918/) (2022)
- [NICE-SLAM: Neural Implicit Scalable Encoding for SLAM](https://mlsanity.com/paper/2112.12130/) (2021)
- [Normalized Object Coordinate Space for Category-Level 6D Object Pose and Size Estimation](https://mlsanity.com/paper/1901.02970/) (2019)
- [SplaTAM: Splat, Track & Map 3D Gaussians for Dense RGB-D SLAM](https://mlsanity.com/paper/2312.02126/) (2023)
- [OmniObject3D: Large-Vocabulary 3D Object Dataset for Realistic Perception, Reconstruction and Generation](https://mlsanity.com/paper/2301.07525/) (2023)
- [What Do Single-view 3D Reconstruction Networks Learn?](https://mlsanity.com/paper/1905.03678/) (2019)
- [SAM 3D: 3Dfy Anything in Images](https://mlsanity.com/paper/2511.16624/) (2025)
- [Native and Compact Structured Latents for 3D Generation](https://mlsanity.com/paper/2512.14692/) (2025)
- [MIDI: Multi-Instance Diffusion for Single Image to 3D Scene Generation](https://mlsanity.com/paper/2412.03558/) (2024)
- [HomebrewedDB: RGB-D Dataset for 6D Pose Estimation of 3D Objects](https://mlsanity.com/paper/1904.03167/) (2019)
- [CAST: Component-Aligned 3D Scene Reconstruction from an RGB Image](https://mlsanity.com/paper/2502.12894/) (2025)
- [Panoptic 3D Scene Reconstruction From a Single RGB Image](https://mlsanity.com/paper/2111.02444/) (2021)
- [vMAP: Vectorised Object Mapping for Neural Field SLAM](https://mlsanity.com/paper/2302.01838/) (2023)
- [NodeSLAM: Neural Object Descriptors for Multi-View Shape Reconstruction](https://mlsanity.com/paper/2004.04485/) (2020)
- [SCube: Instant Large-Scale Scene Reconstruction using VoxSplats](https://mlsanity.com/paper/2410.20030/) (2024)
- [Semantic Scene Completion via Integrating Instances and Scene in-the-Loop](https://mlsanity.com/paper/2104.03640/) (2021)
- [SceneGen: Single-Image 3D Scene Generation in One Feedforward Pass](https://mlsanity.com/paper/2508.15769/) (2025)
- [CoReNet: Coherent 3D scene reconstruction from a single RGB image](https://mlsanity.com/paper/2004.12989/) (2020)
- [PhyRecon: Physically Plausible Neural Scene Reconstruction](https://mlsanity.com/paper/2404.16666/) (2024)
- [PaSCo: Urban 3D Panoptic Scene Completion with Uncertainty Awareness](https://mlsanity.com/paper/2312.02158/) (2023)
- [Physically Compatible 3D Object Modeling from a Single Image](https://mlsanity.com/paper/2405.20510/) (2024)
- [DSP-SLAM: Object Oriented SLAM with Deep Shape Priors](https://mlsanity.com/paper/2108.09481/) (2021)
- [ZeroShape: Regression-based Zero-shot Shape Reconstruction](https://mlsanity.com/paper/2312.14198/) (2023)
- [NViSII: A Scriptable Tool for Photorealistic Image Generation](https://mlsanity.com/paper/2105.13962/) (2021)

## Abstract (arXiv)

Robots operating safely in cluttered everyday environments often need to infer scene geometry from partial observations. Methods that detect objects in 2D and reconstruct them independently struggle in such scenes: a missed object is never reconstructed, a merged detection can fuse two objects, and separately reconstructed meshes may overlap or fail to touch their supporting surfaces. We introduce CODA (Complete Once, Decompose Afterward), a generative model that instead reconstructs the complete scene geometry from a single unsegmented RGB-D image, then separates the surface into the surrounding environment and movable objects. Still, generated scene geometry can drift from the observed partial point cloud. To reduce this drift, CODA uses two explicit 3D grounding mechanisms to keep reconstructed geometry consistent with observed surfaces while completing unseen regions. Experiments on HomebrewedDB and our custom cluttered-scene dataset show more accurate reconstructions and a higher fraction of objects remaining in place under simulated gravity than both object-first and scene-first baselines.