Computer vision
430 agents · most starred first
Computer vision, page 3 of 5
gradio region:us
gradio region:us
gradio region:us
gradio region:us
Real-time Multimodal Spatial Audio-Visual Grounding Engine (inspired by Meta Ray-Ban & Meta Muse). Fuses 2D/3D visual bounding boxes, user head-pose gaze vectors, and acoustic beamforming azimuth angles into cross-modal focal salience scores and resolves deictic spatial references.
Google Gemini Agentic AI Vision for Wildlife Analytics
Rizz glasses (CarletonU Hackathon)
Visiona AI is a real-time assistive system for visually impaired users that detects objects, estimates distance, tracks motion, and predicts collisions using multiple cameras. It provides voice alerts, answers queries, remembers objects (15 min), and supports goal-based assistance.
An experiment in giving an LLM grounded spatial context on a tldraw canvas plus an AI companion that observes, reflects, and proposes changes back, always accept-gated
Python Steamlit web app that takes in a user-specified image URL and context. Then utilizing OpenAI(GPT4), LangChain and SceneXplain, the app generates a suitable caption for that image and context.
Utilizing multiple Gemma 3n agents to analyze webcam footage
Analyze a Video and generate commentary about it with OpenAI's GPT-4V, Text-to-speech, LangChain, Streamlit, Replit, Twilio SendGrid, and OpenCV!
Simple Educational Implementation of OpenAI CLIP in PyTorch
AI-powered media organization add-on for the Virtual Media Folders plugin. Uses vision-capable AI models to analyze actual image content and automatically organize your media library into virtual folders.
This is a teaching material inspired by the Microsoft's challenge project about adding image analysis and generation capabilities to an application.
Python code for openai live video narrator.
OpenAI's CLIP neural network
A WordPress plugin that generates an alt text on upload, or when editing existing images. Uses GPT-5.2 to understand and analyze the images.
A .NET MAUI sample application that uses omni-input LLMs to create "My Pal", a friendly koala that insults you based on a selfie!
A Cloudflare Worker-based API for DocsBot that generates detailed image descriptions and alt text using Google's Gemini AI. This tool processes image URLs and provides concise alt text and comprehensive descriptions, improving image accessibility and searchability.
Build a multimodal AI assistant for smart glasses using Whisper, vision-language models, and LLM hand on code with Google Colab, tested with Meta Ray-Ban Wayfarer.
AI-powered metadata generation for DaVinci Resolve clips using Gemini Vision or OpenAI.
An AI-powered Road Damage Assessment system using YOLOv11s & SAHI to detect potholes and cracks with high precision. Featuring a Streamlit dashboard for real-time analysis, it provides engineering reports based on IRC standards. Integrated with RAG technology to offer expert maintenance consulting derived from civil engineering manuals.
Fashion authentication agent: LangGraph + GPT-4o Vision + RAG. Paste an eBay link, get a confidence score.
Describe images in the WordPress Media Library using a local large language model. Generates titles, captions, descriptions, and alt tags.
gradio region:us
gradio region:us
LLM 驱动的一体化图片对话插件。自动生成多维度图像标签,并通过多模态语义分析,实现智能斗图、主动表情、表情包分享等多种功能,支持 4 种导入图库方法。
AI Camera is a smart image recognition system built using an ESP32-CAM, I2C LCD, and a button trigger. On pressing the button, it captures an image, sends it to OpenAI’s Vision API through an n8n workflow, and displays the identified object(s) on the LCD screen and Serial Monitor. It’s a compact and low-cost AI-powered vision device.
Self-hosted AI security camera platform — multi-camera YOLOv8 detection, InsightFace recognition, and a Qwen 3 14B chat assistant with 19 tools. Docker Compose, zero cloud.
PhotoBooth Chatbot – Capture Moments & Interact with an AI-Powered Photo Assistant
Hesperus: a camera that almost never looks. Describe what you want to know in plain English; a LangGraph tool-using agent plans when to take a photo, a low-power camera (or a phone) captures it, a vision model checks each photo against your goal, and hybrid RAG lets you ask what it saw before. Accepted at BNAIC/BeNeLearn 2026.
YOLO-Agent is an evidence-driven autonomous optimization harness for object detection | YOLO-Agent 是一个面向目标检测的证据驱动自主优化 Harness
Pixel Perfect for harnessing image comparison. The result looks close, but the spacing is off and now you need to explain what to fix in plain english to your agent, good luck.
PhotoMagic:利用 LLM 视觉模型分析 RAW 照片,自动生成后期参数,并通过 RawTherapee 批量输出高质量 JPEG。附带 AI 照片评价、多维度雷达图与前后对比功能。
A video understanding and research agent that can reason over long-form videos
1000-class Chinese vehicle recognition with ResNet-50, deployed to K230D edge board (28 FPS offline) + ReAct agent
Track anything in streaming with Grounding DINO, SAM 2, and LLM
An LLM based food recognizer web app which tell about details about food preset in image, calculates calories, fats etc
Google glass for the blind
Sentiment Analysis app with React-Native
Deploy an API built with Streamlit on Heroku, in order to classify images.
AI apps using OpenAI Vision model.
Agentic anime detection workbench: mock/Ultralytics backends, CLI/API/GUI, tool-calling agent.
gradio smolagents agent smolagent tool agent-course region:us
gradio gradio-custom-component gradio-5 laban-movement-analysis pose-visualisation pose-estimation movement-visualisation visualisation
gradio region:us
streamlit region:us
Resigned Yann-LeCun
Private, fully on-device AI photo search for Android. Ask grounded questions across photos, videos, screenshots, PDFs, people, OCR, places, and events.
This project acts as a proof of concept of the utility of facial expressions of emotion as a modality. Its intention is to explore the entire process of creating useful applications based on this modality to uncover challenges and determine their viability.
This Python script processes images in a specified folder, sends them to the OpenAI API and saves the responses as text files.
NutriScan is a Discord bot that provides nutritional estimates for meals based on images you provide. It utilizes OpenAI's GPT-4 Vision model to analyze images of food and return a concise breakdown of their nutritional content.
A concept face recognition featured demo of virtual police station where people even in remote places with internet access can get help while using it as home security for potential criminals or missing people are detected by your existing home surveillance system.
Agentic Vision-Language Assistant for Earth observation. SatQuery AI fuses Optical and SAR satellite imagery using a deterministic LLM router for visual grounding, bi-temporal change detection, and VQA through natural language.
👔 FitCheck.AI is your personal AI stylist. Upload outfits for savage critiques, auto-tag your wardrobe, and get smart recommendations - powered by Streamlit, LangChain, MongoDB, and VLMs like CLIP and Qwen.
Private in-browser editor for screenshots, photos and PDFs: font-matched text replacement, OCR, redaction, translation and code export. Designed and built by NIKHIL CHARY SRIRAMOJU.
On-device AI camera lens: detects objects, explains what they are used for, and chats with Groq/Gemini. PWA with WebGPU, TensorFlow.js, voice and multilingual replies.
LAM (Locate Anything Model) — standalone prompt based image segmentation + geospatial tiling as a service.
面向基坑施工安全隐患识别的多模态智能检测框架,融合 YOLO 目标检测、Qwen 视觉语言推理、RAG 知识检索与 AST 空间拓扑规则引擎,实现工程场景中的安全要素感知、语义推理、规则判定与可解释隐患输出。
Open-source vision stack with stereo camera hardware, GPU processing, and AI agent for training video classifiers.
Cas d'usage application vision de reconnaissance de vêtements modèle CNN (référence dans la reconnaissance) + recommandations de vêtements à la demande.
A non-contact LLM-driven reading assistant with device-edge-cloud collaboration
AI screen capture for Mac. Drag any region, get instant analysis from Claude, ChatGPT, or Gemini. Download below ↓
A powerful browser extension that provides instant AI assistance directly in your browser. Highlight text, capture screen areas with OCR, or use voice input to get AI-powered answers without leaving your current tab.
LiveVision AI: Real-Time Visual Conversations with Gemini
gradio agent-demo-track region:us
gradio agent-demo-track region:us
gradio mcp-server region:us
gradio region:us
gradio region:us
gradio region:us
gradio region:us
An open-source AI photo advisor that uses a RAG pipeline with GPT-4o Vision and LangChain to provide expert composition and lighting critiques.
EmotiWeave: correctable multimodal emotion recognition with continuous affect tracking and conflict detection.
爱丽丝的图片助手:AstrBot 图片找图、精准挑图与以图搜图功能合集
MEMO, a visual memory assistant that prioritizes useful uncertainty over confident guesses.
AI claims triage for car damage: custom-trained vision models as tools for an auditable LLM agent. Models measure, the agent reasons, rules decide.
YOLO real-time object detection dashboard with MCP integration
A project exploring agentic AI for Visual Question Answering. Compares zero-shot VLM performance (Transformers) against advanced pipelines using OpenCV (tool-use) and Chain-of-Thought (reasoning).
Local AI stack that observes a single space: currently runs YOLO for detection, Qwen-VL for vision captions, Whisper for audio, and Llama for reasoning. Logs to SQLite, builds summaries/analytics, and exposes an OpenAI-style chat API.
A framework utilizing Vertex AI and Gemini for deep, agentic video analytics and insights.
A Python library for measurement, observation, and calibrated judgment from camera images — the perception layer for human-helper AI applications.
gradio region:us
gradio region:us
gradio region:us
gradio anomaly-detection vision-language-model industrial-inspection comparison-aware region:us
gradio region:us
streamlit region:us
transformers pytorch ChatUniVi text-generation video-text-to-text arxiv:2311.08046
Локальный модуль машинного зрения для LLM-агентов на чистом Rust: метрический 3D scene-graph каждый кадр, MCP-сервер, $0/CPU/приватно
Voice-driven extended reality. GPT Realtime hears you, GPT-5.6 Sol plans agentic, and Reframe segments the object and rebuilds the background from monocular depth and TSDF fusion without LiDAR.
gradio region:us
Matched Outcomes, Divergent Gaze: How Foveated MLLMs Search Compared to Humans
Open-source sports video analysis and basketball analytics for Claude Code, Codex and AI agents. Video to coaching reports, clips, narration and explained MP4s; player/ball tracking, shot charts, training plans and optional OpenHoopVision, WASB, SAHI and KaliCalib evaluation. Any sport via plugins.
gradio region:us
gradio region:us
gradio region:us
gradio region:us