Singapore, Singapore

Jiuxiang Gu


Average Co-Inventor Count = 4.1

ph-index = 1

Forward Citations = 9(Granted Patents)


Location History:

  • Singapore, SG (2022)
  • College Park, MD (US) (2023)

Company Filing History:


Years Active: 2022-2025

Loading Chart...
3 patents (USPTO):

Title: Innovations of Jiuxiang Gu: Pioneering Semantic Scene Graphs and Neural Network Distillation

Introduction

Jiuxiang Gu is a prominent inventor based in Singapore, known for his significant contributions to the fields of image processing and artificial intelligence. With a total of three patents to his name, Gu has demonstrated exceptional creativity and technical prowess in the development of advanced methods for generating semantic scene graphs and enhancing neural network learning.

Latest Patents

Gu's latest patents include groundbreaking work in two major areas:

1. **Generating Scene Graphs from Digital Images Using External Knowledge**: This patent discloses innovative methods, systems, and non-transitory computer-readable storage media for generating semantic scene graphs for digital images. By utilizing an external knowledge base for feature refinement, the system identifies object and subgraph proposals, highlighting potential relationships between objects in a digital image. Through this process, it extracts relationships from the external knowledge base, enhancing feature accuracy for both object and subgraph proposals. Subsequently, a semantic scene graph is generated, which can be further refined using the reconstructed image based on object labels.

2. **Knowledge Distillation for Neural Networks Using Multiple Augmentation Strategies**: In this patent, Gu presents a novel approach to efficiently learning parameters of a distilled neural network from a source neural network. The invention employs various strategies for augmenting digital images, enabling it to successfully transfer learned parameters for improved performance. By comparing classifications of heavily augmented images produced by both networks, the disclosed system effectively transfers knowledge through a loss function, enhancing the distilled network's learning capability.

Career Highlights

Jiuxiang Gu is currently affiliated with Adobe Inc., where he continues to push the boundaries of research in image processing and artificial intelligence. His dedication to innovation is reflected in his patent portfolio, which combines sophisticated techniques with practical applications in the tech industry.

Collaborations

Throughout his career, Gu has collaborated with notable coworkers, including Zhe Lin and Handong Zhao. Their collective expertise has contributed to the success of various projects, fostering an environment of creative collaboration and technological advancement.

Conclusion

Jiuxiang Gu stands out as a leading inventor in the realm of artificial intelligence and image processing. His patents on generating semantic scene graphs and optimizing neural network distillation showcase his commitment to pushing the envelope in technology. As he continues to innovate at Adobe Inc., the impact of his work is sure to resonate across multiple industries, paving the way for future developments in these transformative fields.

This text is generated by artificial intelligence and may not be accurate.
Please report any incorrect information to support@idiyas.com
Loading…