# Image-to-Text

This is a statically quantized version based on the Qari-OCR-0.3-SNAPSHOT-VL-2B-Instruct-merged model, mainly used for image-to-text conversion tasks.

Vintern-1B-v3_5 is a 1-billion-parameter vision-language model supporting image-text generation tasks.

MixTex base_ZhEn is an image-to-text model supporting both Chinese and English, released under the MIT License.

Sarashina2-Vision-8B is a large Japanese vision-language model trained by SB Intuitions, based on the Sarashina2-7B and Qwen2-VL-7B image encoders, achieving excellent performance in multiple benchmarks.

A Devanagari optical character recognition model based on the TrOCR architecture, specifically fine-tuned for Nepali/Devanagari script

TrOCR is a Transformer-based OCR model specifically designed for recognizing handwritten mathematical formulas

Florence-2 is an advanced vision foundation model developed by Microsoft, using a prompt-based approach to handle a wide range of vision and vision-language tasks.

Florence-2 is an advanced vision foundation model developed by Microsoft, employing a prompt-based approach to handle a wide range of visual and vision-language tasks.

Donut is a Transformer-based image-to-text model capable of extracting and generating textual content from images.

Libra is a decoupled vision system built upon large language models, possessing fundamental multimodal understanding capabilities.

LLaVA-Phi-3-mini is a fine-tuned LLaVA model based on Phi-3-mini-4k-instruct and CLIP-ViT-Large-patch14-336, specializing in image-to-text tasks.

InfiMM-HD is a high-resolution multimodal model capable of understanding and generating content that combines images and text.

Fine-tuned image-to-text model based on microsoft/git-base

A ViT-GPT2 architecture captcha recognition model fine-tuned on a dataset of 60,000 images, capable of accurately identifying text in captcha images.

Pix2Struct is an image encoder-text decoder model trained for multiple vision-language tasks, including image captioning and visual question answering.

TrOCR is a Korean image-to-text model based on a vision encoder-decoder architecture, using DeiT as the image encoder and RoBERTa as the text decoder.

A multimodal model based on Microsoft's GIT framework, focused on extracting text from student homework images and generating teacher feedback

A Japanese manga text recognition model based on the TrOCR architecture, specifically designed for extracting text content from manga images.

A fine-tuned handwritten OCR model based on Microsoft's trocr-base-handwritten, achieving a character error rate (CER) of 0.0360 on the evaluation set

This is an image captioning model based on the Vision Encoder-Decoder architecture, capable of generating natural language descriptions for input images.

A satellite image caption generation model fine-tuned based on Microsoft GIT-base, generating brief descriptions for NASA Earth Observatory images

Pix2Struct is an image encoder-text decoder model trained on image-text pairs, suitable for various vision-language tasks

Pix2Struct is a vision-language understanding model specifically fine-tuned for scientific chart visual question answering (VQA) tasks

Pix2Struct is an image encoder-text decoder model trained on various image-text pairs for tasks including image captioning and visual question answering.

GIT is a Transformer decoder conditioned on CLIP image tokens and text tokens, designed for tasks like image and video caption generation and visual question answering.

Invoice processing model fine-tuned based on naver-clova-ix/donut-base

A vision-language model trained on the Flickr8k dataset, capable of generating natural language descriptions for input images

An image-to-text model based on ViT and GPT2 architectures, capable of generating reasonable English descriptions for input images

TrOCR is a Transformer-based optical character recognition model specifically designed for handwritten text recognition, fine-tuned on the IAM dataset.