Inventor

TARIQ SARAH

US45 patents

⚠️ This page may combine multiple inventors who share the name “TARIQ SARAH”. Patents are grouped by organization below to help distinguish them — per-person disambiguation is on the roadmap.

TARIQ SARAH

ZOOX INC

Object velocity from images

Hyper planning based on object and/or region

Multi-task learning for real-time semantic and/or depth aware instance segmentation and/or three-dimensional object bounding

Bounding box embedding for object identifying

Machine-learned model training for pedestrian attribute and gesture detection

Object or person attribute characterization

Image embedding for object tracking

Detector-tracker architecture

Tracking objects using sensor data segmentations and/or representations

Pedestrian attribute and gesture detection

Map consistency checker

Object trajectory from wheel direction

Object or person attribute characterization

Hierarchical machine-learning network architecture

Instance segmentation inferred from machine learning model output

Energy consumption control systems and methods for vehicles

Low variance region detection for improved high variance region detection using machine learning

Restricted multi-scale inference for machine learning

Verifying reliability of data used for autonomous driving

Key point detection

Object trajectory from wheel direction

Multi-task learning for real-time semantic and/or depth aware instance segmentation and/or three-dimensional object bounding

System to detect impacts

Training a machine-learned model to detect low variance regions

Emergency vehicle detection and response

Sensor degradation detection and remediation

Machine learning techniques

Map consistency checker

Machine-learned model training for pedestrian attribute and gesture detection

Object velocity from images

Motion prediction based on appearance

Adjusting vehicle models based on environmental conditions

Image embedding for object tracking

Vision architecture

Depth data model training with upsampling, losses, and loss balancing

Emergency vehicle detection and response

Tracking objects using sensor data segmentations and/or representations

Temporal-based deformable kernels

Bounding box embedding for object identifying

Hierarchical machine-learning network architecture

Detector-tracker architecture

Vision architecture

Depth data model training with upsampling, losses and loss balancing

Parallel vehicle model processing

COHEN JONATHAN MICHAEL

Region of interest tracking for fluid simulation