US Patent Application 18298128. ESTIMATION MODEL FOR INTERACTION DETECTION BY A DEVICE simplified abstract

ESTIMATION MODEL FOR INTERACTION DETECTION BY A DEVICE

Organization Name

Samsung Electronics Co., Ltd.

Inventor(s)

Dongfang Zhao of San Diego CA (US)

Yangwen Liang of San Diego CA (US)

Shuangquan Wang of San Diego CA (US)

Kee-Bong Song of San Diego CA (US)

ESTIMATION MODEL FOR INTERACTION DETECTION BY A DEVICE - A simplified explanation of the abstract

This abstract first appeared for US patent application 18298128 titled 'ESTIMATION MODEL FOR INTERACTION DETECTION BY A DEVICE

Simplified Explanation

The patent application describes a method and device for estimating an interaction with a device.

• The method involves configuring a first token and a second token of an estimation model based on the features of a 3D object.
• Different weights are applied to the first and second tokens to produce weighted input tokens.
• An output token is generated by a first encoder layer of the estimation model based on the weighted input tokens.
• The method also includes receiving the first features from a backbone and extracting second features, including 2D features, using a 2D feature extraction model.
• Data generated based on the 2D features is received by the estimation-model encoder.

Original Abstract Submitted

A method and device are disclosed for estimating an interaction with the device. The method includes configuring a first token and a second token of an estimation model according to first features of a 3D object, applying a first weight to the first token to produce a first-weighted input token and applying a second weight that is different from the first weight to the second token to produce a second-weighted input token, and generating, by a first encoder layer of an estimation-model encoder of the estimation model, an output token based on the first-weighted input token and the second-weighted input token. The method may include receiving, at a 2D feature extraction model, the first features from a backbone, extracting, by the 2D feature extraction model, second features including 2D features, and receiving, at the estimation-model encoder, data generated based on the 2D features.