VIGC
Instruction dataset generator
Autonomously generates high-quality image-text instruction fine-tuning datasets
AAAI 2024: Visual Instruction Generation and Correction
90 stars
5 watching
3 forks
Language: Python
last commit: 10 months ago Related projects:
| Repository | Description | Stars |
|---|---|---|
 rucaibox/comvint | Creating synthetic visual reasoning instructions to improve the performance of large language models on image-related tasks | 18 |
| opendatalab/mllm-dataengine | Automates data generation and model training for improving MLLM capabilities | 36 |
 avi-d-coder/implicit-hie | Automates the creation of cabal or stack configuration files for multi-component Haskell projects. | 205 |
 cvlab-columbia/viper | A framework for generating and executing Python code to solve visual inference tasks using large language models | 1,660 |
 dense-analysis/neural | An AI-powered plugin for Vim and Neovim that generates code and performs tasks such as text completion and error checking. | 469 |
 bin123apple/autocoder | An AI model designed to generate and execute code automatically | 814 |
 iamwangyunkai/carla_py | Generates data for CARLA's visual navigation system using raw camera images and instructions. | 8 |
 ncsoft/cap2qa | A dataset and implementation of a method to generate instructions based on visual data | 5 |
 sauci/pydbc | Generates an Abstract Syntax Tree based on DBC-formatted strings | 2 |
 aidc-ai/parrot | A method and toolkit for fine-tuning large language models to perform visual instruction tasks in multiple languages. | 30 |
 h21lab/5gc_build | A tool for generating code from 5G API definitions using OpenAPI generators | 14 |
 lxtgh/omg-seg | Develops an end-to-end model for multiple visual perception and reasoning tasks using a single encoder, decoder, and large language model. | 1,300 |
 ranjaykrishna/visual_genome_python_driver | A Python wrapper providing access to the Visual Genome dataset by downloading and parsing local data | 357 |
 baai-dcai/visual-instruction-tuning | A dataset and model designed to scale visual instruction tuning using language-only GPT-4 models. | 163 |
 taoxugit/attngan | Reproduces text-to-image generation with attentional generative adversarial networks. | 1,339 |