DINOv2: Meta's Self-Supervised Vision Features That Beat Supervised Models
DINOv2 learns visual features from 142 million curated images without labels, producing representations that outperform supervised ImageNet models as frozen feature extractors across classification, segmentation, and depth tasks.
Standard vision model training requires millions of labeled images. DINOv2 from Meta demonstrates that self-supervised learning - using the model's own predictions as supervision - can produce features that surpass supervised training on downstream tasks.
DINOv2 uses self-distillation: a student network learns to match the outputs of a teacher network (exponential moving average of student weights). Both networks see different augmented views of the same image; the student must predict what the teacher sees for the global view. No labels needed.
LVD-142M Dataset
The quality of self-supervised learning depends heavily on training data diversity. DINOv2's LVD-142M (Large-scale Visual Deduplicated) dataset was curated through:
Starting with a large uncurated web image collection
Self-supervised retrieval to find images similar to curated reference datasets
Deduplication to remove near-duplicate images
This produces 142 million diverse, high-quality images without manual annotation - far more than ImageNet's 1.2 million.
Team workspace
Ship faster with chat, meetings, and projects in one place — Zlyqor.
import torch
from transformers import AutoImageProcessor, AutoModel
from PIL import Image
processor = AutoImageProcessor.from_pretrained("facebook/dinov2-large")
model = AutoModel.from_pretrained("facebook/dinov2-large")
model.eval()
image = Image.open("photo.jpg")
inputs = processor(images=image, return_tensors="pt")
with torch.no_grad():
outputs = model(**inputs)
# CLS token: global image representation
cls_features = outputs.last_hidden_state[:, 0, :] # [1, 1024]
# Patch tokens: spatial features for dense tasks
patch_features = outputs.last_hidden_state[:, 1:, :] # [1, num_patches, 1024]
print(f"Global feature shape: {cls_features.shape}")
print(f"Patch feature shape: {patch_features.shape}")
ViT Backbone Variants
Variant
Parameters
Dim
Speed
ViT-S/14
21M
384
Fastest
ViT-B/14
86M
768
Fast
ViT-L/14
307M
1024
Moderate
ViT-g/14
1.1B
1536
Slow
The HuggingFace DINOv2-large page is the standard balance point. ViT-g provides marginal improvements for most use cases.
Fine-Tuning for Custom Vision Tasks
DINOv2 frozen features with a linear head achieve 86.5% top-1 on ImageNet - competitive with many fully supervised models. For custom classification:
import torch.nn as nn
from transformers import AutoModel
class DINOv2Classifier(nn.Module):
def __init__(self, num_classes: int):
super().__init__()
self.backbone = AutoModel.from_pretrained("facebook/dinov2-base")
self.classifier = nn.Linear(768, num_classes)
# Freeze backbone for fast training
for param in self.backbone.parameters():
param.requires_grad = False
def forward(self, pixel_values):
outputs = self.backbone(pixel_values=pixel_values)
cls = outputs.last_hidden_state[:, 0]
return self.classifier(cls)
Training only the linear head converges in minutes and works well with as few as 50 labeled examples per class - DINOv2's features generalize to new visual domains without extensive fine-tuning.
The GitHub repository includes depth estimation and semantic segmentation examples using DINOv2 patch features directly.
Practical deep-dives on LLMs, developer tools, and AI engineering. No filler. Unsubscribe any time.
// written byFIG. AUTH-01
530
Mahmudul Haque Qudrati
CEO & ML Engineer
CEO and ML Engineer at Pristren. Builds AI-powered software for teams and writes about machine learning, LLMs, developer tools, and practical AI applications.
Supervised learning is the most widely used ML paradigm. Here is exactly how the train-measure-adjust loop works, where labels come from, and when the approach breaks down.
