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

Lightweight real-time people counting model based on quantized YOLOX, optimized for edge devices using LiteRT (TensorFlow Lite) .

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

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

HMI system that uses AI to recognize operators during alerts and show personalized troubleshooting instructions.

Detects critical alarm sounds (e.g., fire alarms) and triggers immediate machine shutdown for safety compliance.

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

Real-time activity recognition using motion sensor data from wearable fitness devices.

Deep learning model to detect presentation attacks in biometric systems, enhancing security against spoofing attempts.

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.

High-precision OCR application that detects and transcribes text from real-world images with superior accuracy, ideal for industrial and logistics use cases.