# Does DINOv3 Set a New Medical Vision Standard? Benchmarking 2D and 3D Classification, Segmentation, and Registration

arXiv: https://arxiv.org/abs/2509.06467 · HTML: https://mlsanity.com/paper/2509.06467/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- Builds on 80 of its 112 references that are in the map.
- First posted on arXiv 8 Sep 2025.

## Builds on (most cited first, 30 of 80)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [U-Net: Convolutional Networks for Biomedical Image Segmentation](https://mlsanity.com/paper/1505.04597/) (2015)
- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [A Simple Framework for Contrastive Learning of Visual Representations](https://mlsanity.com/paper/2002.05709/) (2020)
- [Segment Anything](https://mlsanity.com/paper/2304.02643/) (2023)
- [DINOv2: Learning Robust Visual Features without Supervision](https://mlsanity.com/paper/2304.07193/) (2023)
- [Masked Autoencoders Are Scalable Vision Learners](https://mlsanity.com/paper/2111.06377/) (2021)
- [Emerging Properties in Self-Supervised Vision Transformers](https://mlsanity.com/paper/2104.14294/) (2021)
- [Scaling Laws for Neural Language Models](https://mlsanity.com/paper/2001.08361/) (2020)
- [Bootstrap your own latent: A new approach to self-supervised Learning](https://mlsanity.com/paper/2006.07733/) (2020)
- [SAM 2: Segment Anything in Images and Videos](https://mlsanity.com/paper/2408.00714/) (2024)
- [Unsupervised Learning of Visual Features by Contrasting Cluster Assignments](https://mlsanity.com/paper/2006.09882/) (2020)
- [V-Net: Fully Convolutional Neural Networks for Volumetric Medical Image Segmentation](https://mlsanity.com/paper/1606.04797/) (2016)
- [An Empirical Study of Training Self-Supervised Vision Transformers](https://mlsanity.com/paper/2104.02057/) (2021)
- [TransUNet: Transformers Make Strong Encoders for Medical Image Segmentation](https://mlsanity.com/paper/2102.04306/) (2021)
- [3D U-Net: Learning Dense Volumetric Segmentation from Sparse Annotation](https://mlsanity.com/paper/1606.06650/) (2016)
- [DINOv3](https://mlsanity.com/paper/2508.10104/) (2025)
- [Attention-based Deep Multiple Instance Learning](https://mlsanity.com/paper/1802.04712/) (2018)
- [ChestX-ray8: Hospital-scale Chest X-ray Database and Benchmarks on Weakly-Supervised Classification and Localization of Common Thorax Diseases](https://mlsanity.com/paper/1705.02315/) (2017)
- [UNETR: Transformers for 3D Medical Image Segmentation](https://mlsanity.com/paper/2103.10504/) (2021)
- [Swin UNETR: Swin Transformers for Semantic Segmentation of Brain Tumors in MRI Images](https://mlsanity.com/paper/2201.01266/) (2022)
- [Data Efficient and Weakly Supervised Computational Pathology on Whole Slide Images](https://mlsanity.com/paper/2004.09666/) (2020)
- [VoxelMorph: A Learning Framework for Deformable Medical Image Registration](https://mlsanity.com/paper/1809.05231/) (2018)
- [The Medical Segmentation Decathlon](https://mlsanity.com/paper/2106.05735/) (2021)
- [CSWin Transformer: A General Vision Transformer Backbone with Cross-Shaped Windows](https://mlsanity.com/paper/2107.00652/) (2021)
- [ConViT: Improving Vision Transformers with Soft Convolutional Inductive Biases](https://mlsanity.com/paper/2103.10697/) (2021)
- [Self-Supervised Pre-Training of Swin Transformers for 3D Medical Image Analysis](https://mlsanity.com/paper/2111.14791/) (2021)
- [Perception Encoder: The best visual embeddings are not at the output of the network](https://mlsanity.com/paper/2504.13181/) (2025)
- [Deep Learning for Segmentation using an Open Large-Scale Dataset in 2D Echocardiography](https://mlsanity.com/paper/1908.06948/) (2019)
- [Patches Are All You Need?](https://mlsanity.com/paper/2201.09792/) (2022)

## Built on by (most cited first, 7 of 7)

- [Are Vision Foundation Models Foundational for Electron Microscopy Image Segmentation?](https://mlsanity.com/paper/2602.08505/) (2026)
- [Frozen Foundation-Model Embeddings Discard Small-Lesion Signal in Chest Radiography: Implications for Pre-Deployment Evaluation](https://mlsanity.com/paper/2606.11606/) (2026)
- [Prototype Memory-Guided Training-Free Anomaly Classification and Localization in Prenatal Ultrasound](https://mlsanity.com/paper/2607.00744/) (2026)
- [Axial-Centric Cross-Plane Attention for 3D Medical Image Classification](https://mlsanity.com/paper/2602.21636/) (2026)
- [LUMOS: Latent Universal Medical Priors for Segmentation](https://mlsanity.com/paper/2603.01115/) (2026)
- [DINO-3DRA: Leveraging 2D Foundation Model Semantics for 3D Cerebral Aneurysm Segmentation](https://mlsanity.com/paper/2608.07767/) (2026)
- [CMRVision: A Foundation Model for Cardiac MR Image Analysis](https://mlsanity.com/paper/2609.01308/) (2026)

## Abstract (arXiv)

The advent of large-scale vision foundation models, pre-trained on diverse natural images, has marked a paradigm shift in computer vision. However, how the frontier vision foundation models' efficacies transfer to specialised domains such as medical imaging remains an open question. This report investigates whether DINOv3, a state-of-the-art self-supervised vision transformer (ViT) pre-trained on natural images, can directly serve as a powerful, unified encoder for medical vision tasks without domain-specific fine-tuning. To answer this, we benchmark DINOv3 across common medical vision tasks, including 2D and 3D classification, segmentation, and registration on a wide range of medical imaging modalities. We systematically analyse its scalability by varying model sizes and input image resolutions. Our findings reveal that DINOv3 shows impressive performance and establishes a formidable new baseline. Remarkably, it can even outperform medical-specific foundation models like BiomedCLIP and CT-Net on several tasks, despite being trained solely on natural images. However, we identify clear limitations: The model's features degrade in scenarios requiring deep domain specialisation, such as in whole-slide images (WSIs), electron microscopy (EM), and positron emission tomography (PET). Furthermore, we observe that DINOv3 does not consistently follow the scaling law in the medical domain. Its performance does not reliably increase with larger models or finer feature resolutions, showing diverse scaling behaviours across tasks. Overall, our work establishes DINOv3 as a strong baseline, whose powerful visual features can serve as a robust prior for multiple medical tasks. This opens promising future directions, such as leveraging its features to enforce multiview consistency in 3D reconstruction.