3D AI Model Generator

What is the “Aries” series?

The company name of Astraea Software Co., Ltd. is derived from Astraea, the goddess of justice in Greek mythology.

Based on scientific management methods, we will make unselfish management decisions in accordance with justice and fairness, grow the company, remember the altruistic spirit, and pursue the well-being of all customers, society and employees. ..

Based on that idea, we named our product series Aries, Aries, which is one of the 12 constellations.

What is the “3D AI Model Generator”?

In the manufacturing industry of automobiles, electric appliances, etc., 3D CAD systems are widely used, and design data is stored as 3D shapes. However, the current situation is that there is no AI model for handling 3D data.

Based on the latest CNN research results, Astrier Software has succeeded in convolving 3D shape data for the first time in the world.

In order for more people to use this technology more easily, I thought it was necessary to have a mechanism to effectively utilize the 3D data held by customers and easily create AI models.

Therefore, we have developed and released a 3D AI model Generator, which is a product that allows you to easily create an AI model simply by preparing 3D data.

AI model creation procedure

About the procedure for creating a general AI model

〇AI model development is generally carried out in the process shown on the left.

〇There are the following issues in the development of AI models.

  • It takes time and effort to collect and process data according to the construction and learning of AI models.
  • In particular, specialized knowledge of AI is required for building and learning AI models.
  • It is necessary to continue to strengthen the AI ​​model by adding new learning data and re-learning in response to changes in the environment.
  • The causal relationship between the learning content and the result is unclear.

About AI model creation procedure of 3D AI model Generator

〇3D AI model Generator

You can easily create a trained 3D AI model simply by preparing the existing 3D data and its classification name (label).

〇We can handle the development of AI models as follows.

  • Since existing 3D data is effectively used, the time and labor required for data collection and processing can be reduced.
  • Since the Generator builds and implements the AI ​​model construction / learning environment, specialized knowledge of AI is not required.
  • By simply adding new learning data, re-learning can be performed and the AI ​​model can be continuously strengthened.
  • Since the learned data and the result can be confirmed, it is easy to strengthen the AI ​​model such as increasing the learning data of the part with low evaluation.

3D AI model Generator system configuration

About the management location of learning data

  • Please upload the learning data on S3 of cloud services such as AWS.
  • At that time, please divide the hierarchy for each classification (label).

About saving to managed location

  • Use the AWS S3 bucket that you are using and upload the learning data to S3 using the AWS console.
  • Upload the learning data using our web tool for uploading learning data.

About learning 3D AI model

  • If you want to use your S3 environment, you need to build an access environment to your S3 bucket.
  • The 3D AI model Generator periodically acquires and learns the data on the S3 bucket.
About the release of 3D AI model
  • Trained 3D AI models can be released regularly and used through Web API.

About the use of 3D AI model

  • Call the Web API with a 3D image as an argument.
  • As a result, you can get what is similar to the trained data./li>

Management of learning data

About the structure of the management location (hierarchical structure)

  • Training data is stored in a hierarchical structure for each class.
  • If you build it in an environment like AWS S3 and make it accessible, we will periodically acquire learning data and create a 3D AI model.

Management location example (AWS S3)

  • Inside the bolt bucket (Astraea-bolt-database), there is a folder for the company (Astraea-software).
  • Create a folder with a label name for each class in it.
  • Upload the data (3D data) to be actually learned in that folder.

Web tool for learning data management

We provide web tools for managing 3D data for learning for customers who do not use cloud services such as S3.

  • Inside the bolt bucket (Astraea-bolt-database), there is a folder for the company (Astraea-software)./li>
  • In it, manage folders with label names for each class.
  • Click “Details” to check the 3D data currently being managed (learned).
  • Click “Open Modal” to open the 3D Viewer in a pop-up.
  • You can check the 3D data currently managed.
  • You can select the file to upload by dragging and dropping the file to upload or selecting it in the file selection dialog.
  • Click the “Upload File” button to start uploading the file.
  • You can also check the result of the upload process.
  • If you click “Open Modal”, you can also check the file before uploading in the 3D Viewer.

Use of trained 3D AI model

Usage example of 3D AI model (Web API call)

As mentioned above, you can easily call the trained model you created by using Web API.

Our other product, PLM using 3D AI model, also uses the model created by 3D AI model Generator via Web API.

Other

About system configuration

The main function of the 3D AI model Generator is to download the training data, learn it, and release the trained 3D AI model.

Therefore, it is possible to support a flexible system configuration according to the customer’s environment.

About the cost of 3D AI model

The 3D AI model Generator can support flexible system configurations according to the customer’s environment. It is possible to flexibly respond to requests such as wanting to make a small start to reduce initial investment and evaluating 3D AI models.

If you are interested in a 3D AI model Generator that can easily create a 3D AI model and continuously enhance it, please do not hesitate to ask any questions.
We apologize for the inconvenience, but thank you for your inquiry.