Simplify AI development
AITRIOS streamlines the AI development process with its automated AI development workflow, making it accessible to teams of all skill levels. Our platform is specifically designed for edge devices, ensuring AI models perform optimally without extensive optimization.
Optimize end-to-end AI lifecycle management
Streamlined AI development workflows allow for rapid iteration and expansion, ensuring teams can see real-world results quickly. Optimize the end-to-end AI model lifecycle with integrated MLOps capabilities to build, train, and deploy, and maintain AI models on edge AI devices seamlessly.
Scale edge AI and vision AI solutions
By automating MLOps, AITRIOS makes it easier to retrain models as needed and scale edge AI solutions to meet growing demands - a valuable asset for teams looking to leverage AI to drive innovation and improve efficiency.
Studio Series
Accelerate AI development with Studio Series
Local Studio
- Industrial-grade AI training tool that runs locally on PC or server
- Features a no-code interface
- Designed for quick development of simple AI models
- Requires minimal sample data for training - as few as 50 images
Studio
- Cloud-based AI training service
- Features a no-code interface
- Designed for complex tasks and use cases
- Sustainable and scalable for long-term use cases
AI Services Designed for Edge AI
Pre-Trained AI Templates
AITRIOS offers highly adaptable, pre-trained AI model templates that can be easily fine-tuned for specific applications, enabling teams to achieve high-performance models more quickly.
Flexible AI Tools & Services
For teams of all skill levels
Bring your own model: Converter & Packager
Converter & Packager allows teams to convert existing PyTorch and Keras AI models. This provides a popular option for teams deploying models on edge devices equipped with Sony’s IMX500 intelligent vision sensor.
- Leverage your favorite third-party AI services. Develop an AI model using your preferred service and then use Converter & Packager to run the AI model on an edge device equipped with Sony’s IMX500.
Build your own AI model fast and easy: Local Studio
Local Studio is a software suite that provides full-stack AI development and support capabilities including an industrial grade AI training service designed to help teams build models for simple, small-scale use cases locally on their PC or server. It’s the ideal solution for teams that need a straightforward AI model without an extensive development process and resources.
- Build with easy-to-use, no-code tools – Local Studio guides teams through the process of AI development – no need for a machine vision or AI expert.
- Minimal data required for AI model training - Train a model with dozens, rather than thousands of images collected directly from your operating environment. For certain scenarios, teams can train a model with as few as fifty images with Local Studio.
- AI development at the edge - Eliminate the need for complex cloud infrastructure by managing the entire AI development process, from training with local image sets to deploying and visualizing model results, all at the edge.
- Optimized for industrial use cases - Local Studio makes building and managing vision AI models for a wide range of inspection tasks incredibly easy. Whether you need to detect anomalies, classify objects, or build custom inspection datasets, Local Studio simplifies the process. The intuitive, no-code interface empowers machine operators to build AI models without AI expertise. Quickly and easily configure the touchpoints of models to your machine, integrate both camera and machine control seamlessly using Modbus TCP or Ethernet IP protocols, allowing teams to focus on their specific manufacturing needs. Local Studio also supports the GenICam standard to ensure industrial vision AI applications are compatible and easy to use.
Build complex models leveraging AI templates: Studio
Studio is a powerful cloud-based AI service designed to help teams quickly build robust AI models from end-to-end for complex tasks and use cases. Configure advanced AI models on a single edge device to produce customized AI insights.
- Faster AI development with annotation-less learning: Studio accelerates AI model training by eliminating manual data labeling. Models built with Studio learn from unlabeled data, automatically identifying patterns and relationships. This approach significantly reduces development time and costs, making AI creation more efficient and scalable.
- Real world flexibility with automatic AI domain adaptation: With Studio, models can adapt to new scenarios and environments without complete retraining. This crucial feature allows AI to thrive in ever-changing, real-world scenarios. For instance, a model trained to recognize objects in a warehouse could be adapted to recognize objects in a retail store, ensuring adaptability across diverse applications.
- Accelerate development with AI templates: Jumpstart projects using pre-trained AI templates. These templates can be easily customized for specific use cases across various scenarios. By eliminating the need for data annotation, model training, and IMX500 platform conversion, AI templates significantly simplify vision AI implementation, saving time and resources.