YOLOv8n
PulseAugur coverage of YOLOv8n — every cluster mentioning YOLOv8n across labs, papers, and developer communities, ranked by signal.
3 day(s) with sentiment data
-
Offline AI Camera App Uses Two Models for On-Device Detection and Description
A developer has created an open-source camera application that runs entirely offline on mobile devices, utilizing two distinct AI models. The app employs YOLOv8n for object detection and a fine-tuned 0.8B vision-languag…
-
New DVFS scheduler eliminates thermal throttling for edge AI on passively cooled hardware
Researchers have developed a new Dynamic Voltage and Frequency Scaling (DVFS) scheduler designed to prevent thermal throttling in passively cooled edge devices running deep neural networks. This scheduler utilizes time-…
-
New framework uses YOLO and RT-DETR for depth-aware pothole detection
Researchers have developed a depth-aware pothole detection framework that utilizes RGB-D sensors for improved accuracy. The study compared five different architectures, including YOLOv8n, YOLOv8nSeg, YOLOv9t, RTDETRL, a…
-
New system estimates calories in Bangladeshi street food using AI models
Researchers have developed a vision-based system for estimating the calorie content of Bangladeshi street food, addressing a gap in current Western-centric approaches. The study compared five object detection and segmen…
-
AI framework enhances tracheal anatomy understanding for robotic surgery
Researchers have developed a novel learning-based framework for hierarchical tracheal anatomy understanding, specifically designed for ultrasound-guided robotic surgery. This system integrates a YOLOv8n localization bac…
-
YOLOv8n waste detection model underperforms with synthetic data
A study published on arXiv evaluated the effectiveness of synthetic and derived images in improving a YOLOv8n object detection model for campus waste detection. The research found that none of the tested augmentation co…
-
New system enables real-time aerial person tracking on low-power hardware
Researchers have developed a novel system called EMTS-Det for real-time aerial person tracking on low-power hardware. This five-stage system estimates ego-motion, normalizes frame data, and uses a small, efficient netwo…
-
New geometry augmentation boosts few-shot industrial defect detection
This paper introduces Rough Path Signature-Guided Geometry Augmentation (RPS-GA), a novel method to improve few-shot industrial defect detection. RPS-GA treats Canny edge contours as ordered paths and uses their signatu…
-
New geometry augmentation boosts few-shot industrial defect detection
Researchers have developed a new method called rough path signature-guided geometry augmentation (RPS-GA) to improve few-shot industrial surface defect detection. This technique treats Canny edge contours as planar path…
-
Camera-only robots regain line tracking with novel two-stage recovery system · 2 sources tracked
Researchers have developed a novel two-stage recovery system for camera-only autonomous ground vehicles that can re-establish line tracking when visual guidance is lost. The system first attempts to re-acquire the line …
-
Digital twin framework enhances UAV pavement monitoring in traffic
Researchers have developed a digital twin framework using Unity Technologies to improve pavement monitoring by unmanned aerial vehicles (UAVs) in real-world traffic conditions. This framework integrates a YOLOv8n percep…
-
New LYNRED-MDS dataset targets low-visibility pedestrian detection
Researchers have introduced the LYNRED Mobility Dataset Multimodal Detection Subset (LYNRED-MDS), a new dataset designed to improve early collision prediction in low-visibility driving conditions. This subset of the LYN…
-
New HiRes method accurately identifies resistor values from images
Researchers have developed HiRes, a novel hierarchical cascaded pipeline for accurately identifying resistor values from images. This method integrates object detection using YOLOv8n, semantic segmentation with UNet++ a…
-
YOLOv8n Model Compiled for Hailo-8L on Raspberry Pi 5
This article details the process of compiling the YOLOv8n model into the HEF format for use with the Hailo-8L chip on a Raspberry Pi 5. It focuses on preparing the YOLO detection model for deployment on this compact AI …
-
Rio3.5 Model Outperforms Qwen3.7; YOLOv8n Achieves High FPS
The city government of Rio de Janeiro has developed a model named Rio3.5 that has outperformed Qwen3.7 in recent benchmarks. Separately, a project demonstrates dual YOLOv8n UAV detection on an RK3588S chip, achieving 42…
-
AI mobile guide for Grand Egyptian Museum developed
Researchers have developed TimeLens, an AI-powered mobile guide for the Grand Egyptian Museum. This system can recognize artifacts in real-time and answer visitor questions in English or Arabic. The project involved cre…
-
YOLO-AMC enhances building crack detection with attention mechanisms
Researchers have developed YOLO-AMC, an enhanced YOLO architecture designed for improved building crack detection. This model integrates various attention mechanisms, such as GAM, Res-CBAM, and SA, into its feature fusi…
-
New GNN improves multi-object tracking in drone imagery
Researchers have developed HDST-GNN, a novel graph neural network designed for multi-object tracking in UAV aerial imagery. This system addresses challenges like varying altitudes, small and occluded objects, and freque…
-
Generative AI enhances hand detection for safety applications
Researchers have developed a method to improve hand detection models for safety-critical applications by using generative AI to create synthetic data. This synthetic data, which includes variations like gloves and tatto…
-
Sixfab launches AI HAT+ for Raspberry Pi 5, offering 25 TOPS
Sixfab has released the AI HAT+, an NPU accelerator for the Raspberry Pi 5, priced around $100. This HAT connects via PCIe and GPIO, offering a simple setup that recognizes the NPU automatically within 15 minutes. It su…