Data processing device and data processing method

Patent metadata

Field	Value
Publication number	US-12535321-B2
Application number	US-202318238024-A
Country	US
Kind code	B2
Filing date	Aug 25, 2023
Priority date	Aug 31, 2022
Publication date	Jan 27, 2026
Grant date	Jan 27, 2026

Abstract

A data processing device and method that detect a location and direction of a vehicle, recognize an action by an occupant of the vehicle in the form of pointing outside the vehicle, estimate a likely designated region indicated by the occupant that is based on the recognition results and the vehicle location, recognize a target object candidate from the designated region indicated by the occupant; and output information regarding the target object candidate. The size of the designated region is differentiated based on the relationship between the direction pointed at by the occupant and the direction of the vehicle.

First claim

The invention claimed is:

1 . A data processing device installed in a vehicle, comprising:
a vehicle location detecting unit that detects a location and a direction of the vehicle and provides a signal; an occupant recognition unit that recognizes an action of an occupant of the vehicle pointing to outside of the vehicle and provides a signal; an estimation unit coupled to the vehicle location detecting unit and the occupant recognition unit, the estimation unit estimates a likely designated region indicated by the occupant based on the recognition results of the occupant recognition unit and the vehicle location; an object recognition unit that recognizes a target object candidate from the designated region indicated by the occupant; and an output unit that outputs information regarding the target object candidate; wherein the estimation unit
estimates a first region as the designated region based on the direction pointed in by the occupant when the direction pointed in by the occupant is forwards, based on the direction of the vehicle, and
estimates a second region as the designated region based on the direction pointed in by the occupant when the direction pointed in by the occupant is backwards, based on the direction of the vehicle, and
wherein the second region is larger than the first region which is estimated as the designated region when the direction pointed in by the occupant is forwards, based on the direction of the vehicle.

2 . The data processing device according to claim 1 , wherein,
the object recognition unit presents multiple candidates to the occupant when there are a multiple target object candidates in the designated region, and the output unit outputs information regarding the candidate selected by the occupant out of the multiple candidates.

3 . The data processing device according to claim 1 , wherein,
the estimation unit increases the size of the designated region as the speed of the vehicle increases when the direction pointing to by the occupant is backwards.

4 . The data processing device according to claim 1 , wherein,
the estimation unit sets the road surface of a right backward side within a predetermined distance as the designated region when the occupant points to the right backward side, based on the direction of the vehicle, and the estimation unit sets the road surface of a left backward side within a predetermined distance as the designated region when the occupant points to the left backward side, based on the direction of the vehicle.

5 . The data processing device according to claim 4 , wherein,
the estimation unit increases the predetermined distance as the speed of the vehicle increases.

6 . The data processing device according to claim 4 , wherein,
the output unit outputs a video signal including an image captured when a positional relation between the vehicle and the target object candidate becomes a predetermined condition after the object recognition unit recognizes the target object candidate.

7 . The data processing device according to claim 1 , further comprising:
a voice recognition unit for recognizing words included in a voice of the occupant; wherein, the object recognition unit recognizes a target object candidate located in the designated region having a feature indicated by the words.

8 . A data processing method executed in a data processing device installed in a vehicle, comprising the steps of:
detecting a location and direction of the vehicle and providing a signal representative of the vehicle location and the vehicle direction; recognizing an action of an occupant of the vehicle pointing outside the vehicle and providing a signal representative of the recognition results; estimating a likely designated region indicated by the occupant based on the recognition results in the signal provided during the recognizing step and the vehicle location in the signal provided during the detecting step; recognizing a target object candidate from the designated region indicated by the occupant; and outputting information regarding the target object candidate; wherein the estimating step
estimates a first region as the designated region based on the direction pointed in by the occupant when the direction pointed in by the occupant is forwards, based on the vehicle direction, and
estimates a second region as the designated region based on the direction pointed in by the occupant when the direction pointed in by the occupant is backwards, based on the vehicle direction, and
wherein the second region is larger than the first region which is estimated as the designated region when the direction pointed in by the occupant is forwards, based on the vehicle direction.

Assignees

• Faurecia Clarion Electronics Co Ltd

Inventors

• Matsushita Yuki

Classifications

• G10L15/22

  Procedures used during a speech recognition process, e.g. man-machine dialogue · CPC title

• G06V40/20

  Movements or behaviour, e.g. gesture recognition (recognition of facial expressions G06V40/16) · CPC title

• G06V20/59

  inside of a vehicle, e.g. relating to seat occupancy, driver state or inner lighting conditions · CPC title

• G06V20/58

  Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads · CPC title

• G06V20/41

  Higher-level, semantic clustering, classification or understanding of video scenes, e.g. detection, labelling or Markovian modelling of sport events or news items (segmenting video sequences G06V20/49) · CPC title