logo

For the Inventor, By the Inventor

Santa Clara, CA, United States of America

Vladlen Koltun

USPTO Granted Patents = 18 

 

Average Co-Inventor Count = 2.9

ph-index = 2

Forward Citations = 18(Granted Patents)


Company Filing History:

goldMedal16 out of 54,750 
 
Intel Corporation
 patents
silverMedal1 out of 29,302 
 
Robert Boschgmbh
 patents
where one patent can have more than one assignee

Years Active: 2019-2025

Loading Chart...
Loading Chart...
18 patents (USPTO):Explore Patents
1. 12400346 - Methods and apparatus for metric depth estimation using a monocular visual-inertial system (2025-08-26),
2. 12380714 - Methods and apparatus to perform dense prediction using transformer blocks (2025-08-05),
3. 12307579 - Deep geometric model fitting (2025-05-20),
4. 12260630 - Methods and apparatus to implement parallel architectures for neural network classifiers (2025-03-25),
5. 12260483 - Learning neural reflectance shaders from images (2025-03-25),
6. 12153416 - Mobile communication terminal device operation of robot terminal (2024-11-26),
7. 12094026 - Apparatus and method for enhancing graphics rendering photorealism (2024-09-17),
8. 11972519 - Learning neural reflectance shaders from images (2024-04-30),
9. 11928787 - Deep novel view synthesis from unstructured input (2024-03-12),
10. 11875252 - Neural network including a neural network projection layer configured for a summing parameter (2024-01-16),
11. 11816784 - Deep geometric model fitting (2023-11-14),
12. 11610129 - System and method for multiscale deep equilibrium models (2023-03-21),
13. 11393160 - Deep geometric model fitting (2022-07-19),
14. 10803565 - Low-light imaging using trained convolutional neural networks (2020-10-13),
15. 10726858 - Neural network for speech denoising trained with deep feature losses (2020-07-28),

Title: The Innovative Mind of Vladlen Koltun: Revolutionizing Neural Networks

Introduction: Vladlen Koltun, an influential inventor based in Santa Clara, California, has made significant contributions to the field of neural networks and geometric modeling. With a remarkable portfolio of 16 patents, he showcases his commitment to advancing technological innovations that enhance data processing capabilities.

Latest Patents: Among his latest patents, one titled "Deep Geometric Model Fitting" presents a sophisticated system that employs two neural networks to optimize and refine model weights. This innovative technology enables the generation of a geometric model from input data, significantly reducing dependency on the volume of data points. Another notable patent, "Methods and Apparatus to Implement Parallel Architectures for Neural Network Classifiers," outlines a framework for processing images using parallel neural network blocks, enabling enhanced classification accuracy through advanced data fusion and pooling techniques.

Career Highlights: Vladlen has had an impressive career working with prominent companies such as Intel Corporation and Robert Bosch GmbH. His tenure at these companies has allowed him to explore cutting-edge technologies, pushing the boundaries of what is possible within the realms of artificial intelligence and machine learning.

Collaborations: Throughout his career, Vladlen has collaborated with esteemed colleagues, including Qifeng Chen and Stephan Richter. These partnerships have fostered a creative environment that encourages innovative thinking and problem-solving in complex computational challenges.

Conclusion: Vladlen Koltun stands out as a pioneering inventor whose work continues to shape the future of neural networks and geometric modeling. His patents reflect a deep commitment to improving technology and tackling the challenges associated with data processing. As the field of artificial intelligence evolves, Vladlen's contributions will undoubtedly play a crucial role in driving further innovations and advancements.

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