# DAViD: Data-efficient and Accurate Vision Models from Synthetic Data

arXiv: https://arxiv.org/abs/2507.15365 · HTML: https://mlsanity.com/paper/2507.15365/

## Key facts

- Cited by 11 papers in the map (data through 14 Oct 2026).
- Builds on 37 of its 51 references that are in the map.
- First posted on arXiv 21 Jul 2025.

## Builds on (most cited first, 30 of 37)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [DINOv2: Learning Robust Visual Features without Supervision](https://mlsanity.com/paper/2304.07193/) (2023)
- [SGDR: Stochastic Gradient Descent with Warm Restarts](https://mlsanity.com/paper/1608.03983/) (2016)
- [The Cityscapes Dataset for Semantic Urban Scene Understanding](https://mlsanity.com/paper/1604.01685/) (2016)
- [Vision Transformers for Dense Prediction](https://mlsanity.com/paper/2103.13413/) (2021)
- [Expressive Body Capture: 3D Hands, Face, and Body from a Single Image](https://mlsanity.com/paper/1904.05866/) (2019)
- [Towards Robust Monocular Depth Estimation: Mixing Datasets for Zero-shot Cross-dataset Transfer](https://mlsanity.com/paper/1907.01341/) (2019)
- [Depth Anything V2](https://mlsanity.com/paper/2406.09414/) (2024)
- [AMASS: Archive of Motion Capture as Surface Shapes](https://mlsanity.com/paper/1904.03278/) (2019)
- [Generalised Dice overlap as a deep learning loss function for highly unbalanced segmentations](https://mlsanity.com/paper/1707.03237/) (2017)
- [PIFuHD: Multi-Level Pixel-Aligned Implicit Function for High-Resolution 3D Human Digitization](https://mlsanity.com/paper/2004.00452/) (2020)
- [Depth Pro: Sharp Monocular Metric Depth in Less Than a Second](https://mlsanity.com/paper/2410.02073/) (2024)
- [Omnidata: A Scalable Pipeline for Making Multi-Task Mid-Level Vision Datasets from 3D Scans](https://mlsanity.com/paper/2110.04994/) (2021)
- [WIDER FACE: A Face Detection Benchmark](https://mlsanity.com/paper/1511.06523/) (2015)
- [ICON: Implicit Clothed humans Obtained from Normals](https://mlsanity.com/paper/2112.09127/) (2021)
- [Function4D: Real-time Human Volumetric Capture from Very Sparse Consumer RGBD Sensors](https://mlsanity.com/paper/2105.01859/) (2021)
- [BEDLAM: A Synthetic Dataset of Bodies Exhibiting Detailed Lifelike Animated Motion](https://mlsanity.com/paper/2306.16940/) (2023)
- [Sapiens: Foundation for Human Vision Models](https://mlsanity.com/paper/2408.12569/) (2024)
- [AGORA: Avatars in Geography Optimized for Regression Analysis](https://mlsanity.com/paper/2104.14643/) (2021)
- [Fake It Till You Make It: Face analysis in the wild using synthetic data alone](https://mlsanity.com/paper/2109.15102/) (2021)
- [ECON: Explicit Clothed humans Optimized via Normal integration](https://mlsanity.com/paper/2212.07422/) (2022)
- [Whole-Body Human Pose Estimation in the Wild](https://mlsanity.com/paper/2007.11858/) (2020)
- [Learning to Regress 3D Face Shape and Expression from an Image without 3D Supervision](https://mlsanity.com/paper/1905.06817/) (2019)
- [Real-Time High-Resolution Background Matting](https://mlsanity.com/paper/2012.07810/) (2020)
- [MODNet: Real-Time Trimap-Free Portrait Matting via Objective Decomposition](https://mlsanity.com/paper/2011.11961/) (2020)
- [Hi4D: 4D Instance Segmentation of Close Human Interaction](https://mlsanity.com/paper/2303.15380/) (2023)
- [Context-Aware Image Matting for Simultaneous Foreground and Alpha Estimation](https://mlsanity.com/paper/1909.09725/) (2019)
- [High-fidelity 3D Human Digitization from Single 2K Resolution Images](https://mlsanity.com/paper/2303.15108/) (2023)

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

- [From Frames to Sequences: Temporally Consistent Human-Centric Dense Prediction](https://mlsanity.com/paper/2602.01661/) (2026)
- [When Pretty Isn't Useful: Investigating Why Modern Text-to-Image Models Fail as Reliable Training Data Generators](https://mlsanity.com/paper/2602.19946/) (2026)
- [THFM: A Unified Video Foundation Model for 4D Human Perception and Beyond](https://mlsanity.com/paper/2603.25892/) (2026)
- [VideoMaMa: Mask-Guided Video Matting via Generative Prior](https://mlsanity.com/paper/2601.14255/) (2026)
- [Sapiens2](https://mlsanity.com/paper/2604.21681/) (2026)
- [Condition Matters in Full-head 3D GANs](https://mlsanity.com/paper/2602.07198/) (2026)
- [Face Anything: 4D Face Reconstruction from Any Image Sequence](https://mlsanity.com/paper/2604.19702/) (2026)
- [Skullptor: High Fidelity 3D Head Reconstruction in Seconds with Multi-View Normal Prediction](https://mlsanity.com/paper/2602.21100/) (2026)
- [ExpPortrait: Expressive Portrait Generation via Personalized Representation](https://mlsanity.com/paper/2602.19900/) (2026)
- [GeoDiff4D: Geometry-Aware Diffusion for 4D Head Avatar Reconstruction](https://mlsanity.com/paper/2602.24161/) (2026)
- [Revisiting Shape from Polarization in the Era of Vision Foundation Models](https://mlsanity.com/paper/2603.04817/) (2026)

## Abstract (arXiv)

The state of the art in human-centric computer vision achieves high accuracy and robustness across a diverse range of tasks. The most effective models in this domain have billions of parameters, thus requiring extremely large datasets, expensive training regimes, and compute-intensive inference. In this paper, we demonstrate that it is possible to train models on much smaller but high-fidelity synthetic datasets, with no loss in accuracy and higher efficiency. Using synthetic training data provides us with excellent levels of detail and perfect labels, while providing strong guarantees for data provenance, usage rights, and user consent. Procedural data synthesis also provides us with explicit control on data diversity, that we can use to address unfairness in the models we train. Extensive quantitative assessment on real input images demonstrates accuracy of our models on three dense prediction tasks: depth estimation, surface normal estimation, and soft foreground segmentation. Our models require only a fraction of the cost of training and inference when compared with foundational models of similar accuracy. Our human-centric synthetic dataset and trained models are available at https://aka.ms/DAViD.