# NormalCrafter: Learning Temporally Consistent Normals from Video Diffusion Priors

arXiv: https://arxiv.org/abs/2504.11427 · HTML: https://mlsanity.com/paper/2504.11427/

## Key facts

- Cited by 25 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 30 of its 40 references that are in the map.
- First posted on arXiv 15 Apr 2025.

## 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 30)

- [Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2006.11239/) (2020)
- [High-Resolution Image Synthesis with Latent Diffusion Models](https://mlsanity.com/paper/2112.10752/) (2021)
- [Score-Based Generative Modeling through Stochastic Differential Equations](https://mlsanity.com/paper/2011.13456/) (2020)
- [Deep Unsupervised Learning using Nonequilibrium Thermodynamics](https://mlsanity.com/paper/1503.03585/) (2015)
- [Emerging Properties in Self-Supervised Vision Transformers](https://mlsanity.com/paper/2104.14294/) (2021)
- [ScanNet: Richly-annotated 3D Reconstructions of Indoor Scenes](https://mlsanity.com/paper/1702.04405/) (2017)
- [Stable Video Diffusion: Scaling Latent Video Diffusion Models to Large Datasets](https://mlsanity.com/paper/2311.15127/) (2023)
- [Align your Latents: High-Resolution Video Synthesis with Latent Diffusion Models](https://mlsanity.com/paper/2304.08818/) (2023)
- [The Replica Dataset: A Digital Replica of Indoor Spaces](https://mlsanity.com/paper/1906.05797/) (2019)
- [Objaverse-XL: A Universe of 10M+ 3D Objects](https://mlsanity.com/paper/2307.05663/) (2023)
- [Representation Alignment for Generation: Training Diffusion Transformers Is Easier Than You Think](https://mlsanity.com/paper/2410.06940/) (2024)
- [Hypersim: A Photorealistic Synthetic Dataset for Holistic Indoor Scene Understanding](https://mlsanity.com/paper/2011.02523/) (2020)
- [Sora: A Review on Background, Technology, Limitations, and Opportunities of Large Vision Models](https://mlsanity.com/paper/2402.17177/) (2024)
- [Omnidata: A Scalable Pipeline for Making Multi-Task Mid-Level Vision Datasets from 3D Scans](https://mlsanity.com/paper/2110.04994/) (2021)
- [GeoWizard: Unleashing the Diffusion Priors for 3D Geometry Estimation from a Single Image](https://mlsanity.com/paper/2403.12013/) (2024)
- [DepthCrafter: Generating Consistent Long Depth Sequences for Open-world Videos](https://mlsanity.com/paper/2409.02095/) (2024)
- [MatrixCity: A Large-scale City Dataset for City-scale Neural Rendering and Beyond](https://mlsanity.com/paper/2309.16553/) (2023)
- [Lotus: Diffusion-based Visual Foundation Model for High-quality Dense Prediction](https://mlsanity.com/paper/2409.18124/) (2024)
- [StableNormal: Reducing Diffusion Variance for Stable and Sharp Normal](https://mlsanity.com/paper/2406.16864/) (2024)
- [Evaluation of CNN-based Single-Image Depth Estimation Methods](https://mlsanity.com/paper/1805.01328/) (2018)
- [3D Ken Burns Effect from a Single Image](https://mlsanity.com/paper/1909.05483/) (2019)
- [Rethinking Inductive Biases for Surface Normal Estimation](https://mlsanity.com/paper/2403.00712/) (2024)
- [3D Common Corruptions and Data Augmentation](https://mlsanity.com/paper/2203.01441/) (2022)
- [Estimating and Exploiting the Aleatoric Uncertainty in Surface Normal Estimation](https://mlsanity.com/paper/2109.09881/) (2021)
- [Learning Temporally Consistent Video Depth from Video Diffusion Priors](https://mlsanity.com/paper/2406.01493/) (2024)
- [The 2019 DAVIS Challenge on VOS: Unsupervised Multi-Object Segmentation](https://mlsanity.com/paper/1905.00737/) (2019)
- [Fine-Tuning Image-Conditional Diffusion Models is Easier than You Think](https://mlsanity.com/paper/2409.11355/) (2024)
- [What Matters When Repurposing Diffusion Models for General Dense Perception Tasks?](https://mlsanity.com/paper/2403.06090/) (2024)
- [Surface Normal Estimation of Tilted Images via Spatial Rectifier](https://mlsanity.com/paper/2007.09264/) (2020)
- [Buffer Anytime: Zero-Shot Video Depth and Normal from Image Priors](https://mlsanity.com/paper/2411.17249/) (2024)

## Built on by (most cited first, 25 of 25)

- [Video Generation Models are General-Purpose Vision Learners](https://mlsanity.com/paper/2607.09024/) (2026)
- [Lumen: Consistent Video Relighting and Harmonious Background Replacement with Video Generative Models](https://mlsanity.com/paper/2508.12945/) (2025)
- [Diffusion Knows Transparency: Repurposing Video Diffusion for Transparent Object Depth and Normal Estimation](https://mlsanity.com/paper/2512.23705/) (2025)
- [Generative World Renderer](https://mlsanity.com/paper/2604.02329/) (2026)
- [From Frames to Sequences: Temporally Consistent Human-Centric Dense Prediction](https://mlsanity.com/paper/2602.01661/) (2026)
- [CtrlVDiff: Controllable Video Generation via Unified Multimodal Video Diffusion](https://mlsanity.com/paper/2511.21129/) (2025)
- [THFM: A Unified Video Foundation Model for 4D Human Perception and Beyond](https://mlsanity.com/paper/2603.25892/) (2026)
- [Vidu S2: Real-Time Interactive, Editable, and Spatial Video Generation](https://mlsanity.com/paper/2609.11638/) (2026)
- [MADrive: Memory-Augmented Driving Scene Modeling](https://mlsanity.com/paper/2506.21520/) (2025)
- [MMPhysVideo: Physically Plausible Video Generation Through Joint RGB-Perception Modeling](https://mlsanity.com/paper/2604.02817/) (2026)
- [UniVidX: A Unified Multimodal Framework for Versatile Video Generation via Diffusion Priors](https://mlsanity.com/paper/2605.00658/) (2026)
- [Towards Consistent Video Geometry Estimation](https://mlsanity.com/paper/2605.30060/) (2026)
- [Monocular Normal Estimation via Shading Sequence Estimation](https://mlsanity.com/paper/2602.09929/) (2026)
- [Generative Perception of Shape and Material from Differential Motion](https://mlsanity.com/paper/2506.02473/) (2025)
- [Relit-LiVE: Relight Video by Jointly Learning Environment Video](https://mlsanity.com/paper/2605.06658/) (2026)
- [Real2SAM2Real: Generative 3D Caches as Complementary Context for Video Diffusion](https://mlsanity.com/paper/2606.00299/) (2026)
- [RealityBridge: Bridging Editable 3D Gaussian Splatting Driving Simulations and Real-World Videos](https://mlsanity.com/paper/2606.16278/) (2026)
- [Spatial Polarization Multiplexing: Single-Shot Invisible Shape and Reflectance Recovery](https://mlsanity.com/paper/2504.13177/) (2025)
- [TraceFlow: Dynamic 3D Reconstruction of Specular Scenes Driven by Ray Tracing](https://mlsanity.com/paper/2512.10095/) (2025)
- [LSA: Localized Semantic Alignment for Enhancing Temporal Consistency in Traffic Video Generation](https://mlsanity.com/paper/2602.05966/) (2026)
- [HumanOrbit: 3D Human Reconstruction as 360deg Orbit Generation](https://mlsanity.com/paper/2602.24148/) (2026)
- [FLAT: Feedforward Latent Triangle Splatting for Geometrically Accurate Scene Generation](https://mlsanity.com/paper/2606.24876/) (2026)
- [Unified Video Dense Prediction from Disjoint Data](https://mlsanity.com/paper/2607.21592/) (2026)
- [TransNormal-2: Geometry-Grounded Rectified Flow with Edge-Aware Decoding for Precise Normal Estimation](https://mlsanity.com/paper/2609.06665/) (2026)
- [AgentGarten: Code Worlds for Evolving Agents](https://mlsanity.com/paper/2610.12374/) (2026)

## Abstract (arXiv)

Surface normal estimation serves as a cornerstone for a spectrum of computer vision applications. While numerous efforts have been devoted to static image scenarios, ensuring temporal coherence in video-based normal estimation remains a formidable challenge. Instead of merely augmenting existing methods with temporal components, we present NormalCrafter to leverage the inherent temporal priors of video diffusion models. To secure high-fidelity normal estimation across sequences, we propose Semantic Feature Regularization (SFR), which aligns diffusion features with semantic cues, encouraging the model to concentrate on the intrinsic semantics of the scene. Moreover, we introduce a two-stage training protocol that leverages both latent and pixel space learning to preserve spatial accuracy while maintaining long temporal context. Extensive evaluations demonstrate the efficacy of our method, showcasing a superior performance in generating temporally consistent normal sequences with intricate details from diverse videos.