Back to results
US 11,109,152 B2Granted
Optimize the audio capture during conference call in cars
AssigneeAmbarella International LP
InventorPier Paolo Porta
Life of this patent
Estimating the term…
Abstract
An apparatus including an interface and a processor. The interface may be configured to receive video frames corresponding to an interior of a vehicle. The processor may be configured to perform video operations on the video frames to detect objects in the video frames, detect one or more passengers based on the objects detected in the video frames, determine a location of each of the passengers detected, determine which of the passengers may be an audio source and generate a control signal in response to the audio source. The control signal may be configured to adjust an input of a microphone towards the audio source. The video operations may be used to determine the location and the audio source.
Claims20
An apparatus comprising:
- an interface configured to receive pixel data corresponding to an interior of a vehicle;
- and
- a processor configured to
- (i) perform computer vision operations on the pixel data arranged as video frames to detect objects in said video frames,
- (ii) detect one or more passengers based on said objects detected in said video frames,
- (iii) determine a location of each of said passengers detected,
- (iv) determine which of said passengers is a first audio source,
- (v) detect a second audio source from said passengers detected,
- (vi) determine when said first audio source has stopped being active;
- and
- (vii) generate a control signal to adjust an input of a microphone towards (a) said first audio source while said first audio source is active and (b) said second audio source after said first audio source has stopped being active,
- wherein (a) said computer vision operations are used to determine said location and said first audio source, (b) said computer vision operations detect said objects and said locations of said objects within said interior of said vehicle by performing feature extraction based on weight values for each of a plurality of visual features that are associated with said objects extracted from said video frames, (c) said weight values are determined in response to an analysis of training data by said processor prior to said feature extraction, (d) said training data comprises results of said computer vision operations performed on reference images acquired using fleet learning, (e) said fleet learning comprises capturing said reference images from a plurality of different vehicles, (f) said reference images are different from said video frames and (g) said results of said computer vision operations performed on said reference images enable said weight values to be updated while said apparatus is in operation.
9 more elements
An apparatus comprising:
- an interface configured to receive pixel data corresponding to an interior of a vehicle;
- and
- a processor configured to (i) perform computer vision operations on said pixel data arranged as video frames to detect objects in said video frames, (ii) detect one or more passengers based on said objects detected in said video frames, (iii) determine a location of each of said passengers detected, (iv) determine which of said passengers is an audio source and (v) generate a control signal in response to said audio source, wherein (a) said control signal is configured to adjust an input of a microphone towards said audio source, (b) said computer vision operations are used to determine said locations and said audio source, (c) said computer vision operations detect said objects by performing feature extraction based on weight values for each of a plurality of visual features that are associated with said objects extracted from said video frames, (d) said weight values are determined in response to an analysis of training data by said processor prior to said feature extraction, (e) said training data comprises results of said computer vision operations performed on reference images acquired using fleet learning, (f) said fleet learning comprises capturing said reference images from a plurality of different vehicles, (g) said reference images are different from said video frames and (h) said results of said computer vision operations performed on said reference images enable said weight values to be updated while said apparatus is in operation.
Dependencies are inferred from the claim text.