Ijepa Vitg16 22k

🚀 I-JEPA Model (Giant, fine-tuned on IN22K)

I-JEPA is a self - supervised learning method. It predicts the representations of part of an image from those of other parts of the same image. This approach avoids relying on pre - specified invariances for hand - crafted data transformations, which can be biased for specific downstream tasks. It also doesn't require the model to fill in pixel - level details, which often lead to learning less semantically meaningful representations.

🚀 Quick Start

✨ Features

I-JEPA is a self - supervised learning method. It predicts image representations without relying on pre - specified invariances for data transformations and without filling in pixel - level details. It has a predictor working in latent space and can be used for image classification or feature extraction.

📦 Installation

No installation steps are provided in the original document, so this section is skipped.

💻 Usage Examples

Basic Usage

Here is how to use this model for image feature extraction:

import requests
from PIL import Image
from torch.nn.functional import cosine_similarity

from transformers import AutoModel, AutoProcessor

url_1 = "http://images.cocodataset.org/val2017/000000039769.jpg"
url_2 = "http://images.cocodataset.org/val2017/000000219578.jpg"
image_1 = Image.open(requests.get(url_1, stream=True).raw)
image_2 = Image.open(requests.get(url_2, stream=True).raw)

model_id = "jmtzt/ijepa_vitg16_22k"
processor = AutoProcessor.from_pretrained(model_id)
model = AutoModel.from_pretrained(model_id)


def infer(image):
    inputs = processor(image, return_tensors="pt")
    outputs = model(**inputs)
    return outputs.last_hidden_state.mean(dim=1)


embed_1 = infer(image_1)
embed_2 = infer(image_2)

similarity = cosine_similarity(embed_1, embed_2)
print(similarity)

📚 Documentation

How does it work?

As opposed to generative methods that have a pixel decoder, I-JEPA has a predictor that makes predictions in latent space. The predictor in I-JEPA can be seen as a primitive (and restricted) world - model that is able to model spatial uncertainty in a static image from a partially observable context. This world model is semantic in the sense that it predicts high - level information about unseen regions in the image, rather than pixel - level details.

We trained a stochastic decoder that maps the I-JEPA predicted representations back in pixel space as sketches. The model correctly captures positional uncertainty and produces high - level object parts with the correct pose (e.g., dog’s head, wolf’s front legs).

Intended uses & limitations

I-JEPA can be used for image classification or feature extraction. This checkpoint in specific is intended for Feature Extraction.

🔧 Technical Details

No technical details section with more than 50 - word specific descriptions are provided in the original document, so this section is skipped.

📄 License

The license for this project is cc - by - nc - 4.0.

BibTeX entry and citation info

If you use I-JEPA or this code in your work, please cite:

@article{assran2023self,
  title={Self-Supervised Learning from Images with a Joint-Embedding Predictive Architecture},
  author={Assran, Mahmoud and Duval, Quentin and Misra, Ishan and Bojanowski, Piotr and Vincent, Pascal and Rabbat, Michael and LeCun, Yann and Ballas, Nicolas},
  journal={arXiv preprint arXiv:2301.08243},
  year={2023}
}