# SC-Diff: 3D Shape Completion with Latent Diffusion Models

arXiv: https://arxiv.org/abs/2403.12470 · HTML: https://mlsanity.com/paper/2403.12470/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- Placed in 3D, point cloud and scene generation, one of 14 directions in Diffusion models.
- Builds on 38 of its 49 references that are in the map.
- First posted on arXiv 19 Mar 2024.

## Placed in

- Diffusion models / [3D, point cloud and scene generation](https://mlsanity.com/diffusion/3d-point-cloud-and-scene-generation/)

## Builds on (most cited first, 30 of 38)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2006.11239/) (2020)
- [U-Net: Convolutional Networks for Biomedical Image Segmentation](https://mlsanity.com/paper/1505.04597/) (2015)
- [High-Resolution Image Synthesis with Latent Diffusion Models](https://mlsanity.com/paper/2112.10752/) (2021)
- [Denoising Diffusion Implicit Models](https://mlsanity.com/paper/2010.02502/) (2020)
- [Hierarchical Text-Conditional Image Generation with CLIP Latents](https://mlsanity.com/paper/2204.06125/) (2022)
- [Adding Conditional Control to Text-to-Image Diffusion Models](https://mlsanity.com/paper/2302.05543/) (2023)
- [Neural Discrete Representation Learning](https://mlsanity.com/paper/1711.00937/) (2017)
- [ScanNet: Richly-annotated 3D Reconstructions of Indoor Scenes](https://mlsanity.com/paper/1702.04405/) (2017)
- [ShapeNet: An Information-Rich 3D Model Repository](https://mlsanity.com/paper/1512.03012/) (2015)
- [DeepSDF: Learning Continuous Signed Distance Functions for Shape Representation](https://mlsanity.com/paper/1901.05103/) (2019)
- [SDEdit: Guided Image Synthesis and Editing with Stochastic Differential Equations](https://mlsanity.com/paper/2108.01073/) (2021)
- [Consistency Models](https://mlsanity.com/paper/2303.01469/) (2023)
- [Generating Diverse High-Fidelity Images with VQ-VAE-2](https://mlsanity.com/paper/1906.00446/) (2019)
- [3D Shape Generation and Completion through Point-Voxel Diffusion](https://mlsanity.com/paper/2104.03670/) (2021)
- [Latent-NeRF for Shape-Guided Generation of 3D Shapes and Textures](https://mlsanity.com/paper/2211.07600/) (2022)
- [PCN: Point Completion Network](https://mlsanity.com/paper/1808.00671/) (2018)
- [Neural RGB-D Surface Reconstruction](https://mlsanity.com/paper/2104.04532/) (2021)
- [SDFusion: Multimodal 3D Shape Completion, Reconstruction, and Generation](https://mlsanity.com/paper/2212.04493/) (2022)
- [Shape Completion using 3D-Encoder-Predictor CNNs and Shape Synthesis](https://mlsanity.com/paper/1612.00101/) (2016)
- [AutoSDF: Shape Priors for 3D Completion, Reconstruction and Generation](https://mlsanity.com/paper/2203.09516/) (2022)
- [DiffRF: Rendering-Guided 3D Radiance Field Diffusion](https://mlsanity.com/paper/2212.01206/) (2022)
- [Total3DUnderstanding: Joint Layout, Object Pose and Mesh Reconstruction for Indoor Scenes from a Single Image](https://mlsanity.com/paper/2002.12212/) (2020)
- [Diffusion-SDF: Conditional Generative Modeling of Signed Distance Functions](https://mlsanity.com/paper/2211.13757/) (2022)
- [ShapeFormer: Transformer-based Shape Completion via Sparse Representation](https://mlsanity.com/paper/2201.10326/) (2022)
- [A Conditional Point Diffusion-Refinement Paradigm for 3D Point Cloud Completion](https://mlsanity.com/paper/2112.03530/) (2021)
- [SG-NN: Sparse Generative Neural Networks for Self-Supervised Scene Completion of RGB-D Scans](https://mlsanity.com/paper/1912.00036/) (2019)
- [Unsupervised 3D Shape Completion through GAN Inversion](https://mlsanity.com/paper/2104.13366/) (2021)
- [Cooperative Holistic Scene Understanding: Unifying 3D Object, Layout, and Camera Pose Estimation](https://mlsanity.com/paper/1810.13049/) (2018)
- [Scan2Mesh: From Unstructured Range Scans to 3D Meshes](https://mlsanity.com/paper/1811.10464/) (2018)

## Built on by (most cited first, 10 of 10)

- [HoloPart: Generative 3D Part Amodal Segmentation](https://mlsanity.com/paper/2504.07943/) (2025)
- [Points-to-3D: Structure-Aware 3D Generation with Point Cloud Priors](https://mlsanity.com/paper/2603.18782/) (2026)
- [Complete Gaussian Splats from a Single Image with Denoising Diffusion Models](https://mlsanity.com/paper/2508.21542/) (2025)
- [Surf2CT: Cascaded 3D Flow Matching Models for Torso 3D CT Synthesis from Skin Surface](https://mlsanity.com/paper/2505.22511/) (2025)
- [Seen2Scene: Completing Realistic 3D Scenes with Visibility-Guided Flow](https://mlsanity.com/paper/2603.28548/) (2026)
- [TouchAnything: Diffusion-Guided 3D Reconstruction from Sparse Robot Touches](https://mlsanity.com/paper/2604.08945/) (2026)
- [3D Shape Completion with Test-Time Training](https://mlsanity.com/paper/2410.18668/) (2024)
- [OCCDiff: Occupancy Diffusion Model for High-Fidelity 3D Building Reconstruction from Noisy Point Clouds](https://mlsanity.com/paper/2512.08506/) (2025)
- [DinoComplete: 3D Shape Completion with Distilled Semantic Priors and State Space Models](https://mlsanity.com/paper/2605.26949/) (2026)
- [Learning Spherical Occupancy Profiles for Multi-View 3D Reconstruction and Generation](https://mlsanity.com/paper/2608.23206/) (2026)

## Abstract (arXiv)

We present a novel 3D shape completion framework that unifies multimodal conditioning, leveraging both 2D images and 3D partial scans through a latent diffusion model. Shapes are represented as Truncated Signed Distance Functions (TSDFs) and encoded into a discrete latent space jointly supervised by 2D and 3D cues, enabling efficient high-resolution processing while reducing GPU memory usage by 30% compared to state-of-the-art methods. Our approach guides the generation process with flexible multimodal conditioning, ensuring consistent integration of 2D and 3D information from encoding to reconstruction. Our training strategy simulates realistic partial observations, avoiding assumptions about input structure and improving robustness in real-world scenarios. Leveraging our efficient latent space and multimodal conditioning, our model generalizes across object categories, outperforming class-specific models by 12% and class-agnostic models by 47% in l1 reconstruction error, while producing more diverse, realistic, and high-fidelity completions than prior approaches.