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.
Source: BLIP: Bootstrapping Language-Image Pre-training for Unified Vision-Language Understanding and GenerationPaper | Code | Results | Date | Stars |
---|
Task | Papers | Share |
---|---|---|
Retrieval | 13 | 11.40% |
Image Captioning | 10 | 8.77% |
Question Answering | 8 | 7.02% |
Language Modelling | 7 | 6.14% |
Visual Question Answering | 7 | 6.14% |
Visual Question Answering (VQA) | 6 | 5.26% |
Image Generation | 6 | 5.26% |
Image-text matching | 4 | 3.51% |
Video Retrieval | 4 | 3.51% |
Component | Type |
|
---|---|---|
🤖 No Components Found | You can add them if they exist; e.g. Mask R-CNN uses RoIAlign |