Using image caption models to extract prompts in ComfyUI

This is an innovative project aimed at enhancing the visual experience for individuals with impairments. Leveraging machine learning and natural language processing, this repository houses the codebase for generating efficient and coherent natural language descriptions of captured images. The project integrates seamlessly with image recognition,

Image captioning project.

[CVPR23] A cascaded diffusion captioning model with a novel semantic-conditional diffusion process that upgrades conventional diffusion model with additional semantic prior.

Karpathy Splits json files for image captioning

Image captioning model with Resnet50 encoder and LSTM decoder

BLIP-ImageCaption

An Image captioning web application combines the power of React.js for front-end, Flask and Node.js for back-end, utilizing the MERN stack. Users can upload images and instantly receive automatic captions. Authenticated users have access to extra features like translating captions and text-to-speech functionality.

Allows to generate image caption information from annotations

CLIPxGPT Captioner is Image Captioning Model based on OpenAI's CLIP and GPT-2.

A subset of Google's ConceptualCaptions(3M) dataset which include 940k samples.

This repository reimplements "Show, Attend and Tell" model and add extra deep learning techniques.

Major Project Repository

Blip 2 Captioning, Mass Captioning, Question Answering, and other tools.

A Mindspore Implementation of "Show, Attend and Tell: Neural Image Caption Generation with Visual Attention".

A Mindspore Implementation of paper "Show and Tell : Neural Image Caption Generation"

Marshall: Modality-Agnostic Representation learning by SHAred pre-training of muLtiple modaLities

PyTorch implementation of image captioning based on attention mechanism

VisualGPT, CVPR 2022 Proceeding, GPT as a decoder for vision-language models