YOLO

Top Builders

Explore the top contributors showcasing the highest number of app submissions within our community.

YOLO

YOLO (You Only Look Once) is a state-of-the-art, real-time object detection algorithm that can quickly detect and locate objects within an image or video. The YOLO architecture works by taking an input and separating it into a grid of cells and each of these cells is in charge of detecting objects within that region. YOLO returns the bounding boxes containing all the objects in the image and predicts the probability of an object being in each of the boxes and also predicts a class probability to help identify the type of object it is. YOLO is a highly effective object detection algorithm and making YOLO and open-source project led the community to make several improvements in such a limited time.

General
Relese date	2015
Author	Joseph Redmon, Santosh Divvala, Ross Girshick, and Ali Farhadi
Paper	(https://arxiv.org/abs/1506.02640)
Type	Object detection algorithm

YOLO - Resources

Learn even more about YOLO!

v7 Labs Blog "YOLO: Algorithm for Object Detection Explained".
YOLOv5 Repository Object detection architectures and models pretrained on the COCO dataset.
YOLOv6 Web demo Gradio demo for YOLOv6 for object detection on videos.
Hugging Face Spaces Test YOLOv7 in the browser with Hugging Face Spaces.

Edit on GitHub

YOLO AI Technologies Hackathon projects

Discover innovative solutions crafted with YOLO AI Technologies, developed by our community members during our engaging hackathons.

SafeGuard Sentinel

Autonomous robots and AI agents are becoming increasingly common in warehouses, hospitals, construction sites, and public spaces. Yet most systems allow these agents to act freely, with no real-time oversight layer between intent and execution. SafeGuard Sentinel solves this. SafeGuard Sentinel is an AI governance layer that intercepts every proposed robot action before it executes. Using YOLOv8 computer vision, it analyzes the live environment to detect humans and obstacles. A rule-based safety policy engine then evaluates the action against 8 safety rules checking human proximity, speed limits, zone boundaries, and fleet-wide conflict and assigns a risk score from 0 to 100%. Every decision returns one of three verdicts: ALLOW, WARN, or BLOCK. An optional LLM reasoning layer then generates a plain-English explanation of why the decision was made, making the system fully explainable and auditable. Three advanced features make SafeGuard Sentinel production-realistic. Zone Mapping lets operators define restricted, warning, and safe areas directly on the camera feed actions near restricted zones automatically receive elevated risk scores. Multi-Robot Fleet Management tracks multiple agents simultaneously, with fleet-level rules that pause all movement when multiple robots are blocked. The Human Override Panel allows authorized operators to challenge any blocked action within a 2-minute window, with mandatory justification logged to a permanent audit trail. SafeGuard Sentinel demonstrates that autonomous systems don't have to choose between capability and safety. With the right governance layer, every action can be fast, explainable, and human-supervised.

Lucen Robotics

We are building a simulation-first autonomous retail robotics system for night-shift restocking and inventory management. Using NVIDIA Isaac Sim, we model realistic supermarket environments where robots detect out-of-stock shelves, navigate store aisles, and execute pick-and-place restocking tasks autonomously. Our perception pipeline combines computer vision models trained on real supermarket data — for out-of-stock detection and 84.25% SKU classification across 207 products — with depth sensing and SLAM-based navigation within the simulated environment. Robots interact with shelf objects to perform concrete restocking tasks including picking products from backroom inventory, transporting them to target aisles, and placing items in correct shelf positions. The system demonstrates repeatable task execution under varying store layouts and stock conditions, with clear performance metrics and basic failure recovery. A web-based operator dashboard provides real-time monitoring, inventory alerts, and analytics, delivering a complete product experience for retail operators.

NeuroPilot

People who could benefit from hands-free control, for example those with limited mobility, often cannot access it. NeuroPilot changes that: open-source BCI middleware that turns a Muse 2 headband into a real control interface. Muse 2 is already connected to the app, and we have a working integration with DJI Tello. Users train in a 3D simulation and in a training environment; when their brainwave pattern matches a chosen action (move left, right, up, down), the backend fires configurable webhooks or WebSockets. We accumulate 5 actions from the headband; YOLOv8 runs on the Tello camera feed to detect obstacles and Tello state. That batch and the vision data are sent to Gemini, which decides the movement. Every command sent to the DJI Tello is the result of an AI API call, so BCI intent is refined by vision and reasoning before it reaches the drone. The stack is software-only and simulation-first: a Next.js frontend, a FastAPI + WebSocket backend, a web Lab (Three.js), a machine/Connect dashboard for status and live feedback, and per-control webhook URLs. No lab hardware or locked-in vendors. Teams can integrate with existing sims, digital twins, or robots via HTTP.

Havelo AI

Lunar missions cannot rely on human labor. Havelo AI is a fully autonomous robotic system designed to build infrastructure and plant crops on the Moon — entirely in simulation. Our AI uses simulated sensor data to navigate harsh terrain, adapt to environmental changes, assemble modular habitats, and deploy agricultural pods without human intervention. While we demonstrate this on the Moon, our real-world application is autonomous infrastructure deployment for mining, disaster zones, and extreme environments on Earth. Havelo AI is not just a robotics demo — it's a blueprint for autonomous civilization expansion.

Data infrastructure for Real World Robot Training

Robots fail outside the lab because their training data doesn’t transfer. We’re building a data infrastructure platform that converts real-world human demonstrations into reusable robot training data using consumer AR/VR headsets and guided capture workflows. Each demonstration is validated and enriched in a closed loop with action annotations, occlusion handling, and simulation-based verification. Instead of storing isolated trajectories, we decompose demonstrations into spatially and temporally structured atomic actions — reusable skill patterns that generalize across tasks and robots. The resulting skills are exported in industry-standard formats (LeRobot, ManiSkill, Isaac Sim) and can be retargeted across robot embodiments without custom engineering. Because XR devices are widely available, data collection can scale beyond robotics labs to everyday users.

RIFFAI

RIFFAI’s hackathon prototype demonstrates an end-to-end pipeline for energy site intelligence. The prototype ingests multi-source satellite imagery (Google Earth Engine + direct providers), weather feeds and customer training sets; applies an ultra-fast image/geo-metadata alignment pipeline; runs geospatial segmentation (U-Net/CNN) and time-series forecasting (LSTM); and produces explainable suitability scores and a ranked list of candidate polygons for renewable energy deployment. A minimal React web UI or API visualizes results and emits a sample monitoring alert (thermal anomaly or vegetation/water change). The goal is to show that fast metadata processing + energy-tuned models produce practical, developer-ready outputs in near-real time. We’ll provide domain guidance, sample tiles and a small labeled set for teams that request it.