Xiaoyuan Guo

Title: Xiaoyuan Guo: Agricultural Data and Visual Recognition Innovations.

Introduction

Xiaoyuan Guo is an inventor based in Palo Alto, California, whose work includes machine-learning methods for agricultural analysis and image-based object detection. Guo is associated with Deere & Company and has 2 patents.

Latest Patents

Guo’s recent patent work addresses two applications of machine learning.

“Fusing Multimodal Environmental Data for Agricultural Inference” describes combining different types of agricultural data about a parcel. Modality-specific encoders generate embeddings, which a multimodal fusion model combines into a representation of the parcel. That representation can support agricultural predictions and can also be used in training the fusion model or its encoders.

“Object Detection Based on Text Input That Includes Both Target Object Classes and Target Visual Attributes” describes an approach to image classification and detection that uses both object categories and visual characteristics. Text and image information are represented in a shared embedding space, enabling the system to identify objects matching the specified class and attributes. A localization component can also predict image annotations.

Career Highlights

Guo’s listed inventions apply machine learning to distinct technical challenges: interpreting agricultural data across multiple modalities and detecting objects using descriptive text. Together, these approaches illustrate how learned representations can support both agricultural inference and visual recognition.

Collaborations

Guo’s listed coworkers include Jie Yang and Kezhen Chen. The inventions are associated with Deere & Company.

Conclusion

Xiaoyuan Guo’s patent work combines multimodal agricultural analysis with attribute-aware object detection, reflecting a focus on practical machine-learning applications.