LexDanaLexDanabeta
Inventor

RARES A. AMBRUS

Inventor
US

Overview

66→
Patents
2021–2026
Patent span
Physics
Primary field (CPC G)

Patents· 66 as inventor

Public. Assignee & technology derived from the patent record. Each patent links to its LexDana page.

G06Tprimary
G06TImage data processing or generation, in general27%
G06VImage or video recognition or understanding17%
G06NComputing arrangements based on specific computational models14%
G06FElectric digital data processing8%
B60WConjoint control of vehicle sub-units of different type or different function7%
G01SRadio direction-finding7%
G05DSystems for controlling or regulating non-electric variables6%
H04NPictorial communication, e.g. television4%
G06KGraphical data reading3%
G01BMeasuring length, thickness or similar linear dimensions3%
PatentTitleYear
12,567,170Producing a depth map from two-dimensional images202612,530,835Self-supervised depth for volumetric rendering regularization202612,524,952Cross-attention decoding for volumetric rendering202612,511,910Self extrinsic self-calibration via geometrically consistent self-supervised depth and ego-motion learning202512,488,483Geometric 3D augmentations for transformer architectures202512,488,477Systems and methods for multi-object tracking202512,469,152System and method for three-dimensional multi-object tracking202512,430,840Systems and methods for depth synthesis with transformer architectures202512,380,582Warping depth features for depth estimation202512,347,125Using histograms for self-supervised depth estimation202512,340,530Photometric cost volumes for self-supervised depth estimation202512,333,750Systems and methods for generic visual odometry using learned features via neural camera models202512,307,695System and method for self-supervised monocular ground-plane extraction202512,307,694Self-supervised monocular depth estimation via rigid-motion embeddings202512,293,548Systems and methods for estimating scaled maps by sampling representations from a learning model202512,243,260Producing a depth map from a monocular two-dimensional image202512,175,708Systems and methods for self-supervised learning of camera intrinsic parameters from a sequence of images202412,175,775Producing a bird's eye view image from a two dimensional image202412,151,361Robust and perceptive manipulation through soft-bubble grippers and monocular depth202412,073,580Self-supervised 3D keypoint learning for monocular visual odometry202412,067,785Evaluation process for a multi-task network202412,020,489Network architecture for monocular depth estimation and object detection202412,008,818Systems and methods to train a prediction system for depth perception202411,966,234System and method for monocular depth estimation from semantic information202411,948,309Systems and methods for jointly training a machine-learning-based monocular optical flow, depth, and scene flow estimator202411,948,310Systems and methods for jointly training a machine-learning-based monocular optical flow, depth, and scene flow estimator202411,915,487System and method for self-supervised depth and ego-motion overfitting202411,900,626Self-supervised 3D keypoint learning for ego-motion estimation202411,891,094Using a neural network to produce a digital map for a location202411,887,248Systems and methods for reconstructing a scene in three dimensions from a two-dimensional image202411,868,439Mixed-batch training of a multi-task network202411,854,280Learning monocular 3D object detection from 2D semantic keypoint detection202311,822,621Systems and methods for training a machine-learning-based monocular depth estimator202311,798,288System and method for generating a training set for improving monocular object detection202311,783,591Using a multimode neural network to determine a representation of a spatial structure objects in an image202311,775,615System and method for tracking detected objects202311,734,845System and method for self-supervised monocular ground-plane extraction202311,704,822Systems and methods for multi-camera modeling with neural camera networks202311,704,821Camera agnostic depth network202311,663,729Network architecture for the joint learning of monocular depth prediction and completion202311,657,522Sparse auxiliary network for depth completion202311,652,972Systems and methods for self-supervised depth estimation according to an arbitrary camera202311,628,865Method and system for behavioral cloning of autonomous driving policies for safe autonomous agents202311,625,846Systems and methods for training a machine-learning-based monocular depth estimator202311,625,905System and method for tracking occluded objects202311,615,544Systems and methods for end-to-end map building from a video sequence using neural camera models202311,557,051Training of joint depth prediction and completion202311,531,892Systems and methods for detecting and matching keypoints between different views of a scene202211,508,080Systems and methods for generic visual odometry using learned features via neural camera models202211,501,490Systems and methods for multi-resolution fusion of pseudo-LiDAR features202211,494,927Systems and methods for self-supervised depth estimation202211,436,743Systems and methods for semi-supervised depth estimation according to an arbitrary camera202211,430,218Using a bird's eye view feature map, augmented with semantic information, to detect an object in an environment202211,398,043System and method for self-supervised monocular depth regularization from surface normals202211,386,567Systems and methods for weakly supervised training of a model for monocular depth estimation202211,328,517System and method for generating feature space data202211,321,862Systems and methods for multi-camera modeling with neural camera networks202211,321,863Systems and methods for depth estimation using semantic features202211,315,269System and method for generating a point cloud that includes surface normal information202211,257,231Camera agnostic depth network202211,256,986Systems and methods for training a neural keypoint detection network202211,210,802Systems and methods for conditioning training data to avoid learned aberrations202111,176,709Systems and methods for self-supervised scale-aware training of a model for monocular depth estimation202111,144,818Network architecture for ego-motion estimation202111,145,074Systems and methods for depth estimation using monocular images202111,138,751Systems and methods for semi-supervised training using reprojected distance loss2021

Career & activity timeline

Assembled from patent assignee/location history and declaration credentials.

2021–2026Patents assigned to TOYOTA RES INST INC (66)· patent assignee history
2021–2026Based in US· patent location history
2023–2025Patents assigned to UNIV LELAND STANFORD JUNIOR (3)· patent assignee history
2024–2026Patents assigned to MASSACHUSETTS INST TECHNOLOGY (3)· patent assignee history
2024–2025Patents assigned to TOYOTA TECH INSTITUTE AT CHICAGO (3)· patent assignee history
2025–2026Patents assigned to TOYOTA MOTOR CO LTD (13)· patent assignee history

Inventor activity

Assignee and location history derived from the patent record.

Assignees
AssigneePatentsSpan
TOYOTA RES INST INC662021–2026
TOYOTA MOTOR CO LTD132025–2026
MASSACHUSETTS INST TECHNOLOGY32024–2026
TOYOTA TECH INSTITUTE AT CHICAGO32024–2025
UNIV LELAND STANFORD JUNIOR32023–2025
Locations
LocationPatentsSpan
US662021–2026