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

Code

• github.com/salesforce/lavis
  Officialpytorch★ 11,193
• github.com/huggingface/transformers
  pytorch★ 158,292
• github.com/salesforce/blip
  In paperpytorch★ 5,696
• github.com/towhee-io/towhee
  pytorch★ 3,459
• github.com/ninatu/howtocaption
  pytorch★ 58
• github.com/pwc-1/Paper-8/tree/main/blip
  mindspore★ 0
• github.com/2023-MindSpore-4/Code12/tree/main/MindFormers/blip2
  mindspore★ 0
• github.com/alibaba/EasyNLP/tree/master/examples/blip_retrieval
  jax★ 0
• github.com/2024-MindSpore-1/Code2/tree/main/model-1/blip
  mindspore★ 0

Abstract

Vision-Language Pre-training (VLP) has advanced the performance for many vision-language tasks. However, most existing pre-trained models only excel in either understanding-based tasks or generation-based tasks. Furthermore, performance improvement has been largely achieved by scaling up the dataset with noisy image-text pairs collected from the web, which is a suboptimal source of supervision. In this paper, we propose BLIP, a new VLP framework which transfers flexibly to both vision-language understanding and generation tasks. BLIP effectively utilizes the noisy web data by bootstrapping the captions, where a captioner generates synthetic captions and a filter removes the noisy ones. We achieve state-of-the-art results on a wide range of vision-language tasks, such as image-text retrieval (+2.7% in average recall@1), image captioning (+2.8% in CIDEr), and VQA (+1.6% in VQA score). BLIP also demonstrates strong generalization ability when directly transferred to video-language tasks in a zero-shot manner. Code, models, and datasets are released at https://github.com/salesforce/BLIP.

Tasks

Benchmark Results

Reproductions