Awesome

BLIP: Bootstrapping Language-Image Pre-training for Unified Vision-Language Understanding and Generation

Announcement: BLIP is now officially integrated into LAVIS - a one-stop library for language-and-vision research and applications!

This is the PyTorch code of the <a href="https://arxiv.org/abs/2201.12086">BLIP paper</a> [blog]. The code has been tested on PyTorch 1.10. To install the dependencies, run <pre/>pip install -r requirements.txt</pre>

Catalog:

• Inference demo
• Pre-trained and finetuned checkpoints
• Finetuning code for Image-Text Retrieval, Image Captioning, VQA, and NLVR2
• Pre-training code
• Zero-shot video-text retrieval
• Download of bootstrapped pre-training datasets

Inference demo:

Run our interactive demo using Colab notebook (no GPU needed). The demo includes code for:

1. Image captioning
2. Open-ended visual question answering
3. Multimodal / unimodal feature extraction
4. Image-text matching

Try out the Web demo, integrated into Huggingface Spaces 🤗 using Gradio.

Replicate web demo and Docker image is also available at

Pre-trained checkpoints:

Finetuned checkpoints:

Image-Text Retrieval:

1. Download COCO and Flickr30k datasets from the original websites, and set 'image_root' in configs/retrieval_{dataset}.yaml accordingly.
2. To evaluate the finetuned BLIP model on COCO, run:

3. To finetune the pre-trained checkpoint using 8 A100 GPUs, first set 'pretrained' in configs/retrieval_coco.yaml as "https://storage.googleapis.com/sfr-vision-language-research/BLIP/models/model_base.pth". Then run:

Image-Text Captioning:

1. Download COCO and NoCaps datasets from the original websites, and set 'image_root' in configs/caption_coco.yaml and configs/nocaps.yaml accordingly.
2. To evaluate the finetuned BLIP model on COCO, run:

3. To evaluate the finetuned BLIP model on NoCaps, generate results with: (evaluation needs to be performed on official server)

4. To finetune the pre-trained checkpoint using 8 A100 GPUs, first set 'pretrained' in configs/caption_coco.yaml as "https://storage.googleapis.com/sfr-vision-language-research/BLIP/models/model_base_capfilt_large.pth". Then run:

VQA:

1. Download VQA v2 dataset and Visual Genome dataset from the original websites, and set 'vqa_root' and 'vg_root' in configs/vqa.yaml.
2. To evaluate the finetuned BLIP model, generate results with: (evaluation needs to be performed on official server)

3. To finetune the pre-trained checkpoint using 16 A100 GPUs, first set 'pretrained' in configs/vqa.yaml as "https://storage.googleapis.com/sfr-vision-language-research/BLIP/models/model_base_capfilt_large.pth". Then run:

NLVR2:

1. Download NLVR2 dataset from the original websites, and set 'image_root' in configs/nlvr.yaml.
2. To evaluate the finetuned BLIP model, run

3. To finetune the pre-trained checkpoint using 16 A100 GPUs, first set 'pretrained' in configs/nlvr.yaml as "https://storage.googleapis.com/sfr-vision-language-research/BLIP/models/model_base.pth". Then run:

Finetune with ViT-L:

In order to finetune a model with ViT-L, simply change the config file to set 'vit' as large. Batch size and learning rate may also need to be adjusted accordingly (please see the paper's appendix for hyper-parameter details). <a href="https://github.com/facebookresearch/fairscale">Gradient checkpoint</a> can also be activated in the config file to reduce GPU memory usage.

Pre-train:

1. Prepare training json files where each json file contains a list. Each item in the list is a dictonary with two key-value pairs: {'image': path_of_image, 'caption': text_of_image}.
2. In configs/pretrain.yaml, set 'train_file' as the paths for the json files .
3. Pre-train the model using 8 A100 GPUs:

Zero-shot video-text retrieval:

1. Download MSRVTT dataset following the instructions from https://github.com/salesforce/ALPRO, and set 'video_root' accordingly in configs/retrieval_msrvtt.yaml.
2. Install decord with <pre>pip install decord</pre>
3. To perform zero-shot evaluation, run

Pre-training datasets download:

We provide bootstrapped pre-training datasets as json files. Each json file contains a list. Each item in the list is a dictonary with two key-value pairs: {'url': url_of_image, 'caption': text_of_image}.

Citation

If you find this code to be useful for your research, please consider citing.

Acknowledgement

The implementation of BLIP relies on resources from <a href="https://github.com/salesforce/ALBEF">ALBEF</a>, <a href="https://github.com/huggingface/transformers">Huggingface Transformers</a>, and <a href="https://github.com/rwightman/pytorch-image-models/tree/master/timm">timm</a>. We thank the original authors for their open-sourcing.