https://omerkurkutlu.github.io/tags/object-detection/Object DetectionHugoBlox Kit (https://hugoblox.com)en-usWed, 01 Jul 2026 00:00:00 +0000https://omerkurkutlu.github.io/media/icon_hu_eee4a95885829ab2.pnghttps://omerkurkutlu.github.io/tags/object-detection/https://omerkurkutlu.github.io/projects/tinyml-edge-ai/Wed, 01 Jul 2026 00:00:00 +0000https://omerkurkutlu.github.io/projects/tinyml-edge-ai/<h2 id="overview">Overview</h2> <p>This project develops lightweight perception for resource-constrained edge platforms, from Raspberry Pi to microcontrollers. Images are converted into a compact grid-based obstacle representation, and a lightweight temporal smoothing mechanism improves detection consistency under motion blur, occlusion, and lighting changes with virtually no computational overhead.</p> <h2 id="key-features">Key Features</h2> <ul> <li>Temporally consistent, grid-based obstacle detection</li> <li>Embedded inference with YOLOv8n, SSD-MobileNet v2, and FOMO</li> <li>BLE communication and energy-aware sensing</li> <li>Deployment across desktop GPUs, Raspberry Pi, and MCUs</li> </ul> <h2 id="related-publication">Related Publication</h2> <ul> <li><em>Lightweight Temporal Consistency for Grid-Based Obstacle Detection in Edge Devices.</em> O. Kurkutlu and A. Roohi. IEEE ACSOS-C, 2025.</li> </ul>