Pocket-Sized Multimodal AI for content understanding and generation across multilingual texts, images, and 🔜 video, up to 5x faster than OpenAI CLIP and LLaVA 🖼️ & 🖋️

Reproducible scaling laws for contrastive language-image learning (https://arxiv.org/abs/2212.07143)

Using Segment-Anything and CLIP to generate pixel-aligned semantic features.

[Official] [IROS 2024] A goal-oriented planning to lift VLN performance for Closed-Loop Navigation: Simple, Yet Effective

Clipora is a powerful toolkit for fine-tuning OpenCLIP models using Low Rank Adapters (LoRA).

A simple open-sourced SigLIP model finetuned on Genshin Impact's image-text pairs.

Text-to-image search with OpenCLIP, Docker, Flask, Faiss, etc. and a basic front-end.

Mori_Cloud is a web platform that allows users to store, manage, and share memorable moments through images and text. Powered by artificial intelligence for smart image search, Mori_Cloud delivers a personalized, secure, and modern user experience.

use SAM and OpenCLIP to perform zero-shot object detection using COCO 2017 val split.

[GENERIC] API for practical large models

VALORA AI is a Multimodal Pricing Prediction Model that uses textual and visual data to make precise predictions on product prices

Masked Multi-Component Gated Decomposition Architecture

Group images by provided labels using OpenAI/CLIP

This work represents my Master's Degree Thesis in Applied Computer Science (Machine Learning and Big Data)

CLIP based Zero Shot Instance Segmentation

An Edge-AI security pipeline for real-time violence detection and automated forensic reporting using YOLO, CoCa, and Qwen2.5-VL.

Turn any YouTube video into viral clips for free!

An AI-powered computer vision system that automatically selects the best wedding photos from thousands of images.

A redesigned CLIP architecture replacing the ViT encoder with modern CNN backbones (ConvNeXt V2) to improve efficiency and inference speed while maintaining strong vision-language alignment, enabling real-time use in robotics and edge applications.