BGE-VL-v1.5-zs

MegaPairs: Massive Data Synthesis For Universal Multimodal Retrieval#

Build Build Build Build Build Build Build

News#

Release Plan#

• [x] Paper
• [x] BGE-VL-base and BGE-VL-large models
• [x] BGE-VL-MLLM model
• [x] MegaPairs Dataset
• [x] Evaluation code examples
• [ ] Fine-tuning code

Introduction#

In this work, we introduce MegaPairs, a novel data synthesis method that leverages open-domain images to create heterogeneous KNN triplets for universal multimodal retrieval. Our MegaPairs dataset contains over 26 million triplets, and we have trained a series of multimodal retrieval models, BGE-VL, including BGE-VL-CLIP (base and large) and BGE-VL-MLLM.

BGE-VL achieve state-of-the-art performance on four popular zero-shot composed image retrieval benchmarks and the massive multimodal embedding benchmark (MMEB). Extensive experiments demonstrate the efficiency, scalability, and generalization features of MegaPairs. Please refer to our paper for more details.

Model Usage#

Using Sentence Transformers#

Install Sentence Transformers:

pip install sentence_transformers[image]

This model supports text, image, and combined text+image inputs. Images must be resized to 512x512, as the model was trained with that resolution. Pass a prompt with a task instruction to format inputs as queries, or omit it to format as candidates.

import torch
from sentence_transformers import SentenceTransformer
from PIL import Image

model = SentenceTransformer("BAAI/BGE-VL-v1.5-zs", model_kwargs={"torch_dtype": torch.float16})

# Composed image retrieval: text + image query
query_img = Image.open("./assets/cir_query.png").resize((512, 512))
query_emb = model.encode(
    {"text": "Make the background dark, as if the camera has taken the photo at night", "image": query_img},
    prompt="Retrieve the target image that best meets the combined criteria by using both the provided image and the image retrieval instructions: ",
)

# Image-only candidates (no prompt = candidate format)
candidate_imgs = [
    Image.open("./assets/cir_candi_1.png").resize((512, 512)),
    Image.open("./assets/cir_candi_2.png").resize((512, 512)),
]
candidate_embs = model.encode(candidate_imgs, batch_size=1)
print(candidate_embs.shape)
# (2, 4096)

similarities = model.similarity(query_emb, candidate_embs)
print(similarities)
# tensor([[0.3320, 0.1440]], dtype=torch.float16)

1. BGE-VL-CLIP Models#

You can easily use BGE-VL-CLIP models based on transformers

Our code works well on transformers==4.45.2, and we recommend using this version.

import torch
from transformers import AutoModel

MODEL_NAME = "BAAI/BGE-VL-base" # or "BAAI/BGE-VL-large"

model = AutoModel.from_pretrained(MODEL_NAME, trust_remote_code=True) # You must set trust_remote_code=True
model.set_processor(MODEL_NAME)
model.eval()

with torch.no_grad():
    query = model.encode(
        images = "./assets/cir_query.png", 
        text = "Make the background dark, as if the camera has taken the photo at night"
    )

    candidates = model.encode(
        images = ["./assets/cir_candi_1.png", "./assets/cir_candi_2.png"]
    )
    
    scores = query @ candidates.T
print(scores)

See the demo for a complete example of using BGE-VL for multimodel retrieval.

2. BGE-VL-MLLM Models#

Our code works well on transformers==4.45.2, and we recommend using this version.

import torch
from transformers import AutoModel
from PIL import Image

MODEL_NAME= "BAAI/BGE-VL-MLLM-S1"

model = AutoModel.from_pretrained(MODEL_NAME, trust_remote_code=True)
model.eval()
model.cuda()

with torch.no_grad():
    model.set_processor(MODEL_NAME)

    query_inputs = model.data_process(
        text="Make the background dark, as if the camera has taken the photo at night", 
        images="./assets/cir_query.png",
        q_or_c="q",
        task_instruction="Retrieve the target image that best meets the combined criteria by using both the provided image and the image retrieval instructions: "
    )

    candidate_inputs = model.data_process(
        images=["./assets/cir_candi_1.png", "./assets/cir_candi_2.png"],
        q_or_c="c",
    )

    query_embs = model(**query_inputs, output_hidden_states=True)[:, -1, :]
    candi_embs = model(**candidate_inputs, output_hidden_states=True)[:, -1, :]
    
    query_embs = torch.nn.functional.normalize(query_embs, dim=-1)
    candi_embs = torch.nn.functional.normalize(candi_embs, dim=-1)

    scores = torch.matmul(query_embs, candi_embs.T)
print(scores)

MegaPairs Dataset Card#

We are excited to release the MegaPairs dataset on Hugging Face, which contains over 26 million training samples tailored for composed image retrieval and universal multimodal retrieval tasks.

Dataset Structure#

Each entry in the dataset consists of the following fields:

• q_img: str The file path to the query image.
• q_text: list A list of textual query statements related to the query image. During training, you can randomly select one statement from this list.
• t_img: str The file path to the target image, which serves as the positive example for the combination of q_img and q_text.
• hns: list A list of file paths for hard negative sample images. These are challenging distractors that are visually or semantically similar to the query. It is recommended to include at least one hard negative sample during training, with hns[0] (the query image itself) being a mandatory choice. In our experiments, we used four hard negative samples per query.

Usage#

The dataset is available for download and exploration on Hugging Face. We encourage researchers and practitioners to leverage this dataset to advance multimodal retrieval research and systems.

Model Performance#

Zero-Shot Composed Image Retrieval#

BGE-VL sets a new performance benchmark in zero-shot composed image retrieval tasks. On the CIRCO benchmark, our BGE-VL-base model, with only 149 million parameters, surpasses all previous models, including those with 50 times more parameters. Additionally, BGE-VL-MLLM achieves an 8.1% improvement over the previous state-of-the-art model.

Image: image

Zero-Shot Performance on MMEB#

BGE-VL-MLLM achieves state-of-the-art zero-shot performance on the Massive Multimodal Embedding Benchmark (MMEB), despite being trained only on the ImageText-to-Image paradigm. This demonstrates the excellent generalization capability of MegaPairs for multimodal embedding.

Image: image

Fine-Tuning Performance on MMEB#

After fine-tuning on downstream tasks, BGE-VL-MLLM maintains its leading performance. Notably, it surpasses the previous state-of-the-art by 7.1% on the MMEB out-of-distribution (OOD) set. These results demonstrate the robust generalization capability of BGE-VL-MLLM and highlight the potential of MegaPairs as foundational training data for universal multimodal embedding.

Image: image

Performance Scaling#

MegaPairs showcases scalability: BGE-VL-base improves as training data increases. It also demonstrates efficiency: with just 0.5M training samples, BGE-VL-base significantly outperforms MagicLens, which uses the same CLIP-base backbone and was trained on 36.7M samples.

Image: image

License#

The annotations for MegaPairs and the BGE-VL models are released under the MIT License. The images in MegaPairs originate from the Recap-Datacomp, which is released under the CC BY 4.0 license.

Citation#

If you find this repository useful, please consider giving a star ⭐ and citation

@article{zhou2024megapairs,
  title={MegaPairs: Massive Data Synthesis For Universal Multimodal Retrieval},
  author={Zhou, Junjie and Liu, Zheng and Liu, Ze and Xiao, Shitao and Wang, Yueze and Zhao, Bo and Zhang, Chen Jason and Lian, Defu and Xiong, Yongping},
  journal={arXiv preprint arXiv:2412.14475},
  year={2024}
}