{"id":25135030,"url":"https://github.com/headless-start/yolov4-tiny-raspberrypi","last_synced_at":"2026-04-10T01:05:15.424Z","repository":{"id":271892547,"uuid":"914900518","full_name":"headless-start/yolov4-tiny-raspberrypi","owner":"headless-start","description":"This repository contains YOLOv4-Tiny deployment on Raspberry pi.","archived":false,"fork":false,"pushed_at":"2025-02-01T08:30:32.000Z","size":3930,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":1,"default_branch":"main","last_synced_at":"2025-02-08T16:17:41.734Z","etag":null,"topics":["darknet","edge","numpy","object-detection","opencv","powerpoint","python-3","pytorch","raspberry-pi","tensorflowlite","yolov4","yolov4-darknet","yolov4-tiny"],"latest_commit_sha":null,"homepage":"","language":null,"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/headless-start.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","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}},"created_at":"2025-01-10T14:41:31.000Z","updated_at":"2025-02-01T12:56:13.000Z","dependencies_parsed_at":"2025-01-10T15:42:54.495Z","dependency_job_id":"cf7b1513-260e-4bde-8f7d-ec5284c86487","html_url":"https://github.com/headless-start/yolov4-tiny-raspberrypi","commit_stats":null,"previous_names":["headless-start/object_detection_v4tiny","headless-start/yolov4-tiny-raspberrypi"],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/headless-start%2Fyolov4-tiny-raspberrypi","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/headless-start%2Fyolov4-tiny-raspberrypi/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/headless-start%2Fyolov4-tiny-raspberrypi/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/headless-start%2Fyolov4-tiny-raspberrypi/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/headless-start","download_url":"https://codeload.github.com/headless-start/yolov4-tiny-raspberrypi/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":246922246,"owners_count":20855345,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","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":["darknet","edge","numpy","object-detection","opencv","powerpoint","python-3","pytorch","raspberry-pi","tensorflowlite","yolov4","yolov4-darknet","yolov4-tiny"],"created_at":"2025-02-08T16:17:43.271Z","updated_at":"2025-12-30T23:09:57.139Z","avatar_url":"https://github.com/headless-start.png","language":null,"funding_links":[],"categories":[],"sub_categories":[],"readme":"# Real-Time Object Detection on Edge Devices with YOLOv4-Tiny  \n\n## 📌 Project Overview  \nThis project analyzes the suitability of **YOLOv4-Tiny** for real-time object detection on **edge devices** like the Raspberry Pi. The lightweight architecture of YOLOv4-Tiny enables high-speed inference, making it ideal for resource-constrained hardware. This work is presented in the form of a **college task**, focusing on the architectural advantages and performance optimizations of YOLOv4-Tiny over vanilla YOLOv4.  \n\n---\n\n## 🚀 Key Features  \n1. **Lightweight Architecture**:  \n   - YOLOv4-Tiny uses fewer layers and smaller backbones, making it faster and more efficient for edge deployment.  \n2. **Real-Time Performance**:  \n   - Optimized for speed and efficiency while maintaining acceptable accuracy for real-world applications.  \n3. **Edge Device Optimization**:  \n   - Tested on devices like Raspberry Pi 4B for optimal detection.\n\n---\n\n## 🛠 Architecture and Performance Enhancements  \n### YOLOv4-Tiny Architecture  \n1. **Backbone**:  \n   - CSPDarknet53-Tiny for efficient feature extraction.  \n2. **Neck**:  \n   - Feature Pyramid Network (FPN) for feature pooling.  \n3. **Head**:  \n   - Dual-scale detection for large and small objects using 13×13 and 26×26 grids.  \n\n### Performance Enhancements  \n1. **Fewer Parameters**:\n   - ~6 million compared to ~63.6 million in YOLOv4.  \n3. **Simpler Activation**:\n   - Uses LeakyReLU instead of Mish activation for faster computation.  \n5. **Quantization and Pruning**:\n   - Reduces model size and increases speed for edge deployment.  \n7. **TPU Acceleration**:\n   - Compatible with hardware like Google Coral for faster inference.  \n\n---\n\n## 🛠 Usage  \n### Dependencies  \n- **Python**: 3.8+  \n- **Libraries**: `tensorflow-lite` ,`darknet` , `numpy`, `opencv-python`\n- **Hardware**: Raspberry Pi 4B \n\n---\n\n## 📄 License  \nThis project is licensed under the MIT License. See the [LICENSE](LICENSE) file for details.  \n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fheadless-start%2Fyolov4-tiny-raspberrypi","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fheadless-start%2Fyolov4-tiny-raspberrypi","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fheadless-start%2Fyolov4-tiny-raspberrypi/lists"}