Apparatus, NPU and chipset implemented for fusion neural network

A neural processing unit (NPU), comprising:

• an on-chip memory configured to store information related to a fusion neural network model;
• and
• at least one processing element configured to:
• perform operations of the fusion neural network model based on a first input data and a second input data,
• wherein the first input data includes a first data relating to a visible light image,
• wherein the second input data includes a second data other than the visible light image,
• wherein the fusion neural network model is adapted to process both the visible light image and the second data other than the visible light image, and
• wherein a plurality of layers of the fusion neural network model are operated in an order based on the information.

A chipset, comprising:

• an on-chip memory configured to store information related to a fusion neural network model;
• and
• a neural processing unit (NPU) configured to:
• perform operations of the fusion neural network model based on a first input data and a second input data,
• wherein the first input data includes a first data relating to a visible light image,
• wherein the second input data includes a second data other than the visible light image,
• wherein the fusion neural network model is adapted to process both the visible light image and the second data other than the visible light image, and
• wherein a plurality of layers of the fusion neural network model are operated in an order based on the information.