uform3-image-text-multilingual-base

<h1 align="center">UForm</h1> <h3 align="center"> Pocket-Sized Multimodal AI<br/> For Content Understanding and Generation<br/> In Python, JavaScript, and Swift<br/> </h3>

---

The uform3-image-text-multilingual-base UForm model is a tiny vision and multilingual language encoder, covering 21 languages, mapping them into a shared vector space. This model produces up to 256-dimensional embeddings and is made of:

• Text encoder: 12-layer BERT for up to 50 input tokens.
• Visual encoder: ViT-B/16 for images of 224 x 224 resolution.

Unlike most CLIP-like multomodal models, this model shares 4 layers between the text and visual encoder to allow for more data- and parameter-efficient training. Also unlike most models, UForm provides checkpoints compatible with PyTorch, ONNX, and CoreML, covering the absolute majority of AI-capable devices, with pre-quantized weights and inference code. If you need a larger, more accurate, or multilingual model, check our HuggingFace Hub. For more details on running the model, check out the UForm GitHub repository.

Evaluation

For all evaluations, the multimodal part was used unless otherwise stated.

Monolingual

¹ It's important to note, that the MS-COCO train split was present in the training data.

Multilingual

Recall@10 on the XTD-10 dataset:

Recall@1, Recall@5, and Recall@10 on the COCO-SM dataset:

For a deeper comparison of output ranking check the following table for the Normalized Discounted Cumulative Gains for the first 20 results - NDCG@20:

Installation

pip install "uform[torch,onnx]"

Usage

To load the model:

from uform import get_model, Modality

import requests
from io import BytesIO
from PIL import Image

model_name = 'unum-cloud/uform3-image-text-multilingual-base'
modalities = [Modality.TEXT_ENCODER, Modality.IMAGE_ENCODER]
processors, models = get_model(model_name, modalities=modalities)

model_text = models[Modality.TEXT_ENCODER]
model_image = models[Modality.IMAGE_ENCODER]
processor_text = processors[Modality.TEXT_ENCODER]
processor_image = processors[Modality.IMAGE_ENCODER]

To encode the content:

text = 'a cityscape bathed in the warm glow of the sun, with varied architecture and a towering, snow-capped mountain rising majestically in the background'
image_url = 'https://media-cdn.tripadvisor.com/media/photo-s/1b/28/6b/53/lovely-armenia.jpg'
image_url = Image.open(BytesIO(requests.get(image_url).content))

image_data = processor_image(image)
text_data = processor_text(text)
image_features, image_embedding = model_image.encode(image_data, return_features=True)
text_features, text_embedding = model_text.encode(text_data, return_features=True)