AI-assisted HMI that detects operator presence during machine alerts and provides contextual troubleshooting information.

Real-time AI-powered vision system for detecting defects in manufacturing environments

Real-time detection of critical alarm and emergency sounds for industrial safety applications.

Real-time people detection and counting for industrial and monitoring applications.

Lightweight and quantized PoseNet model for real-time multi-person pose estimation on edge hardware.

Real-time semantic segmentation for understanding roads, vehicles, pedestrians, and other elements in visual scenes.

High-resolution neural network for accurate keypoint detection in human pose estimation tasks.

Real-time image classification for recognizing visual categories on edge devices.

Detects and locates multiple object categories in images and video streams using the lightweight YOLOX architecture.

AI-powered image analysis tool for defect detection in industrial environments, with a guided journey for customizing new defect types.

Classifies physical activities in real time from wearable motion sensor data.

Detects face presentation attacks and verifies liveness from RGB images to strengthen biometric authentication systems..

Transcribes spoken audio into text using the Whisper speech recognition model, optimized for edge inference.

Deep learning system to upscale and enhance low-resolution images while preserving fine details.

A time-series forecasting model designed to predict sensor data patterns in smart buildings, enabling proactive energy management and maintenance.

Detects and recognizes text in complex real-world images using a high-accuracy OCR pipeline for edge applications.

Detects and recognizes text in images using a lightweight OCR pipeline optimized for edge devices.

Detects manufacturing defects in PCB images for automated visual quality inspection on edge devices.

Detects faces and performs face recognition using a lightweight Edge Impulse pipeline designed for edge devices.

Identifies or verifies speakers from voice recordings using robust ECAPA-TDNN speaker embeddings.

Advanced transformer-based detector optimized for accurate, real-time edge performance on SECO platforms.

On-device fine-tuning/validation of ResNet classifiers—customize without cloud dependency.

Detects and maps land-cover changes from multi-temporal satellite or aerial imagery for environmental monitoring and urban analysis.

High-accuracy speech transcription optimized for automated documentation and voice-controlled operations in demanding environments.

Energy-efficient, real-time object detection optimized for low-power edge devices with strict battery and thermal constraints.

AI vision application to detect human falls in real time for safety, elderly care, and industrial risk monitoring.

Classifies human actions from video sequences to support activity understanding, workflow analysis, and safety applications.

Transcribes spoken audio into text using Wav2Vec2 for edge-based speech recognition applications.

Energy-efficient object detection optimized for edge platforms with power and thermal constraints.

Bridging the Field and Digital: Universal Connectivity, Zero Latency, and On-Device Machine Learning for Industry 5.0.

Detects and locates objects in images using the efficient EfficientDet architecture optimized for edge deployment.