{"id":35157851,"url":"https://github.com/jonathanscholtes/azure-ai-damage-detection","last_synced_at":"2026-04-25T15:36:00.395Z","repository":{"id":323924391,"uuid":"1095262515","full_name":"jonathanscholtes/azure-ai-damage-detection","owner":"jonathanscholtes","description":"Few-shot Azure AI Agent for Vehicle Damage Detection — using GPT-4o in Azure AI Foundry to identify visible damage and return structured JSON with bounding boxes and confidence scores.","archived":false,"fork":false,"pushed_at":"2025-11-12T22:42:54.000Z","size":24506,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":0,"default_branch":"main","last_synced_at":"2026-04-25T15:35:56.373Z","etag":null,"topics":["agentic-ai","azure-ai-foundry","damage-detection","llm","python","semantic-kernel"],"latest_commit_sha":null,"homepage":"","language":"Jupyter Notebook","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/jonathanscholtes.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE.md","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":null,"maintainers":null,"copyright":null,"agents":null,"dco":null,"cla":null}},"created_at":"2025-11-12T20:15:07.000Z","updated_at":"2025-11-12T22:42:58.000Z","dependencies_parsed_at":null,"dependency_job_id":null,"html_url":"https://github.com/jonathanscholtes/azure-ai-damage-detection","commit_stats":null,"previous_names":["jonathanscholtes/azure-ai-damage-detection"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/jonathanscholtes/azure-ai-damage-detection","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/jonathanscholtes%2Fazure-ai-damage-detection","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/jonathanscholtes%2Fazure-ai-damage-detection/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/jonathanscholtes%2Fazure-ai-damage-detection/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/jonathanscholtes%2Fazure-ai-damage-detection/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/jonathanscholtes","download_url":"https://codeload.github.com/jonathanscholtes/azure-ai-damage-detection/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/jonathanscholtes%2Fazure-ai-damage-detection/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":32267710,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-04-25T09:15:33.318Z","status":"ssl_error","status_checked_at":"2026-04-25T09:15:31.997Z","response_time":59,"last_error":"SSL_read: unexpected eof while reading","robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":false,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["agentic-ai","azure-ai-foundry","damage-detection","llm","python","semantic-kernel"],"created_at":"2025-12-28T17:05:39.101Z","updated_at":"2026-04-25T15:36:00.387Z","avatar_url":"https://github.com/jonathanscholtes.png","language":"Jupyter Notebook","funding_links":[],"categories":[],"sub_categories":[],"readme":"\u003e **⚠️ Active Development Notice**  \n\u003e This project is currently in active development and may contain breaking changes.  \n\u003e Updates and modifications are being made frequently, which may impact stability or functionality.  \n\u003e This notice will be removed once the project reaches a stable release.\n\n# Azure AI Foundry Damage Detection Agent\n\nAn AI-powered **few-shot image comparison agent** built with **Azure AI Foundry** and **Semantic Kernel**.  \nThe agent compares two images of a vehicle (before and after) to detect visible damage such as dents, scratches, and cracks.  \nIt uses **few-shot examples** to improve consistency in reasoning and JSON output when identifying and localizing visible damage.\n\n---\n\n## Overview  \n\nThis project demonstrates how to use **Azure AI Agents** in **Azure AI Foundry** to perform visual reasoning tasks such as image comparison and structured extraction.  \n\nThe solution leverages **few-shot learning** with **Semantic Kernel** to provide the model with pre-labeled examples of vehicle images.  \nThese examples help the model learn how to identify visual differences and structure its JSON responses accurately for new image pairs.\n\nThe agent:\n- Accepts two input images (`before` and `after`)  \n- Detects visible vehicle damage (dents, scratches, cracks, deformations, etc.)  \n- Outputs bounding boxes and confidence scores in **pixel coordinates**  \n- Leverages **few-shot examples** to improve detection accuracy and JSON consistency  \n- Returns structured JSON output that can be easily processed or visualized  \n\n---\n\n![design](/media/detection1.png)\n\n---\n\n## 🛠️ **Core Steps for Solution Implementation**\n\nFollow these key steps to successfully deploy and configure the solution:\n\n### 1️⃣ [**Deploy the Solution**](docs/deployment.md)\n-  Instructions for deploying solution, including prerequisites, configuration steps.  \n\n### 2️⃣ [**Hands-On Notebook: Damage Detection with Semantic Kernel**](docs/notebooks.md)\n- Follow the guided notebook to create and test a few-shot Azure AI Agent that detects and describes vehicle damage using GPT-4o.\n\n\n\n---\n\n## ♻️ **Clean-Up**\n\nAfter completing the workshop and testing, ensure you delete any unused Azure resources or remove the entire Resource Group to avoid additional charges.\n\n---\n\n## 📜 License  \nThis project is licensed under the [MIT License](LICENSE.md), granting permission for commercial and non-commercial use with proper attribution.\n\n---\n\n## Disclaimer  \nThis code and demo application are intended for educational and demonstration purposes. It is provided \"as-is\" without any warranties, and users assume all responsibility for its use.","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fjonathanscholtes%2Fazure-ai-damage-detection","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fjonathanscholtes%2Fazure-ai-damage-detection","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fjonathanscholtes%2Fazure-ai-damage-detection/lists"}