Boris Chidlovskii

This inventor holds 59 USPTO granted patents and 17 published patent applications and 10 EPO patents, primarily in Domain Adaptation (CPC class G06K). Top assignees: Xerox Corporation, Naver Corporation, Conduent Business Services, LLC. Active years: 2001-2026.

USPTO Granted Patents = 59 (Patent Applications = 17)

% Patents Active = 50.8

EPO Patents = 10 (Patent Applications = 15)

Average Co-Inventor Count = 1.5

ph-index = 16

Forward Citations = 1,760(Granted Patents)

Forward Citations (Not Self Cited) = 1,718(Dec 10, 2025)

Inventors with similar research interests:

Company Filing History:

Conduent Business Services, LLC

patents

4. 4 out of 48

Naver Labs Corporation

patents

where one patent can have more than one assignee

Years Active: 2001-2026

Loading Chart...

Areas of Expertise:

Geometric Vision Models

Unsupervised Learning

Depth Estimation

Indoor Positioning

Image Classification

User Behavior Profiling

Multi-Task Learning

Document Annotation

Demand Prediction

Semantic Segmentation

Feature Extraction

Vehicle Classification

59 patents (USPTO):Sort by Forward CitationsExplore Patents

1. 12730447 - Systems and methods for multi-object navigation in real environments (2026-09-08),

2. 12626489 - System and method using pyramidal and uniqueness matching priors for identifying correspondences between images (2026-05-12),

3. 12499678 - Unsupervised pre-training of geometric vision models (2025-12-16),

4. 12217484 - Method of unsupervised domain adaptation in ordinal regression (2025-02-04),

5. 11960294 - Self-supervised attention learning for depth and motion estimation (2024-04-16),

6. 11927448 - Deep smartphone sensors fusion for indoor positioning and tracking (2024-03-12),

7. 11600017 - Adversarial scene adaptation for camera pose regression (2023-03-07),

8. 11494660 - Latent code for unsupervised domain adaptation (2022-11-08),

9. 11373096 - Semi-supervised variational autoencoder for indoor localization (2022-06-28),

10. 11263756 - Method and apparatus for semantic segmentation and depth completion using a convolutional neural network (2022-03-01),

11. 10430736 - System and method for estimating a dynamic origin-destination matrix (2019-10-01),

12. 10354199 - Transductive adaptation of classifiers without source data (2019-07-16),

13. 10296846 - Adapted domain specific class means classifier (2019-05-21),

14. 10289909 - Conditional adaptation network for image classification (2019-05-14),

15. 10089640 - Methods and systems for interpretable user behavior profiling in off-street parking (2018-10-02),

Title: Innovations of Boris Chidlovskii in Unsupervised Domain Adaptation

Introduction

Boris Chidlovskii, based in Meylan, France, is an accomplished inventor renowned for his contributions to the fields of machine learning and computer vision. With an impressive portfolio of 56 patents, Chidlovskii has made significant advancements that improve how we understand and process images.

Latest Patents

Chidlovskii's latest patents showcase his innovative approach to solving complex problems. One notable patent is the "Method of Unsupervised Domain Adaptation in Ordinal Regression." This invention involves the joint training of a transferable feature extractor network, an ordinal regressor network, and an order classifier network, enhancing the capability of machines to learn from unlabeled data effectively. Specifically, it includes a minimax optimization procedure on a set of labeled source images and unlabeled target images, utilizing a domain discriminator network to differentiate between them, and training both the ordinal regressor and order classifier networks for improved accuracy.

Another significant patent is the "Self-Supervised Attention Learning for Depth and Motion Estimation." This system comprises a depth module for generating depth maps from images captured at different times, a pose module for determining camera positions, and a motion module for assessing variations in camera movement. This patent is crucial for applications in robotics and augmented reality, promoting advancements in depth perception and motion analysis.

Career Highlights

Throughout his career, Boris Chidlovskii has held key positions at prominent companies, including Xerox Corporation and Naver Corporation. His work has continued to push the envelope in image processing and artificial intelligence, making him a respected figure in these domains.

Collaborations

Chidlovskii has collaborated with notable professionals, including Gabriela Csurka and Stéphane Clinchant. These partnerships have resulted in groundbreaking research and innovations, reinforcing the importance of teamwork in the technological advancement landscape.

Conclusion

Boris Chidlovskii exemplifies the spirit of innovation with his remarkable contributions to the realms of machine learning and image processing. His extensive patent portfolio and collaborative efforts have not only advanced the field but also inspired a new generation of inventors and researchers to explore the possibilities of technology in enhancing everyday experiences.