The Code allows users to upload an image and generates captions for the uploaded image using the ViT-GPT2 vision encoder-decoder model. It provides an easy-to-use interface for caption generation

A native caption generation application using the Show and Tell model

Image Captioning Using CLIP & GPT Models

FastAPI backend for ML-Apps, a collection of text-summarization, caption-generation and cartoonization services.

Neural Image Captioning with Attention — PyTorch implementation of "Show, Attend and Tell". Generates natural language captions for images using ResNet/VGG encoders + LSTM with attention.

DivStats Div 1 and Div 2 and mBLEU for caption diversity evaluation

Thai Child's Spontaneous Speech Data

Image Caption Generation using Spatial Attention

for web scraping & AI caption generation with Python

Italian Child's Spontaneous Speech Data

A some what optimized implementation of some light weight and popular models

Flask frontend for ML-Apps, a collection of text-summarization, caption-generation and cartoonization services.

Caption generator that allows users to upload or provide links to images. Partly powered by GPT-4o.

Captionify: Describing Images with AI An AI-powered image captioning system that uses CNNs and LSTMs to generate human-like captions for images. Trained on the Flickr8k dataset and evaluated with BLEU scores, it bridges computer vision and natural language processing for real-world applications like accessibility, social media, and e-commerce.

Caption generator using Vision Language Models and vLLM

The code with an attempt on the disentanglement for the decoder part of a caption generator.

CaptionCraft is an innovative tool that leverages generative AI to create catchy, engaging captions for any image. Users can upload an image or provide an image URL, and CaptionCraft will analyze the visual content to generate a tailored caption to make your images stand out.

🖼️BLIP-powered Image-to-Text Generator achieving 136.7 CIDEr score on TextCaps benchmark (142K captions, 28K images). 129M parameter model with batch processing (25 images), ZIP export with embedded captions & conditional captioning. Live Streamlit demo for instant AI-powered captioning.

"Flickr30k_image_captioning" is a project or repository focused on image captioning using the Flickr30k dataset. The project aims to develop and showcase algorithms and models that generate descriptive captions for images.