Systems and methods for determining a depth or reflectance of objects

Patent metadata

Field	Value
Publication number	US-9912861-B1
Application number	US-201615058566-A
Country	US
Kind code	B1
Filing date	Mar 2, 2016
Priority date	Mar 2, 2016
Publication date	Mar 6, 2018
Grant date	Mar 6, 2018

Abstract

Systems, methods, and computer-readable media are disclosed for determining a depth or reflectance of objects. Example methods may include illuminating a scene within a field of view of a device at a first illuminance value, detecting a reflected illuminance value, and determining a first reflectance value for a first object in the scene. Example methods may include identifying the first object, determining an orientation of the first object, and determining an estimated distance between the device and the first object based at least in part on the first illuminance value, the reflected illuminance value, and the first reflectance value.

First claim

What is claimed is:

1. A camera comprising:
a flash device configured to generate artificial light at a first illuminance value;
a lens having a field of view;
a display configured to display visual representation of a scene within the field of view of the lens;
a light sensor;
at least one memory that stores computer-executable instructions; and
at least one processor configured to access the at least one memory and execute the computer-executable instructions to:
determine that a first object in the scene is a human face;
cause the flash device to illuminate the scene;
measure a reflected illuminance value of the scene using the light sensor while the scene is illuminated;
determine a first reflectance value for the human face;
determine that the first reflectance value is within a predetermined tolerance of an expected reflectance value for human faces in front facing orientations, wherein the expected reflectance value represents an estimated reflectance of incident light from human faces;
calculate a first distance between the camera and the human face using the following formula:

DISTANCE
=

(

REFLECT
*

Y
CHAR

*

DIST
CHAR
4

*

ILLUM
TORCH

)

(

Y
*

ILLUM
CHAR

*
18
⁢
%

)

4

wherein ILLUM TORCH is the first illuminance value, Y is the reflected illuminance value, REFLECT is the first reflectance value, Y CHAR is a light sensor brightness obtained at characterization, DIST CHAR is a characterization distance, ILLUM CHAR is an illuminance of the light sensor at characterization, and 18% is a midtone gray target; and
position the lens with respect to the camera to focus on the human face using the first distance.

2. The camera of claim 1 , wherein the at least one processor is further configured to access the at least one memory and execute the computer-executable instructions to:
determine a first lux value of ambient light incident upon the camera; and
determine a difference between a second lux value of the first illuminance value and the first lux value of the ambient light to calculate the reflected illuminance value.

3. The camera of claim 1 , wherein the at least one processor is further configured to access the at least one memory and execute the computer-executable instructions to:
capture an image of the scene, the image comprising at least one pixel;
wherein the at least one processor is configured to determine the first reflectance value by converting a pixel value of the at least one pixel to the first reflectance value.

4. The camera of claim 1 , wherein the human face is a first human face, and the at least one processor is further configured to access the at least one memory and execute the computer-executable instructions to:
detect a second object in the scene;
determine that the second object is a second human face;
determine a second reflectance value for the second human face;
calculate a second distance between the camera and the second human face;
determine an average of the first distance and the second distance; and
adjust the lens of the camera using the average.

5. A method comprising:
causing, by one or more computer processors coupled to at least one memory, illumination of a scene within a field of view of an imaging device, wherein the scene is illuminated at a first illuminance value;
determining a first lux value of ambient light in the scene;
determining a difference between a second lux value of the illumination and the first lux value of the ambient light;
determining a reflected illuminance value based at least in part on the difference;
determining a first reflectance value for a first object in the scene; and
determining an estimated distance between the imaging device and the first object based at least in part on the first illuminance value, the reflected illuminance value, and the first reflectance value.

6. The method of claim 5 , wherein identifying the first object comprises:
determining the first object in the scene;
identifying the first object using an object recognition algorithm;
determining that the first object is a human face in a front facing orientation;
determining that the first reflectance value corresponds to an expected reflectance value for human faces in front facing orientations; and
autofocusing a lens of the imaging device on the human face.

7. The method of claim 5 , further comprising:
capturing an image of the scene, wherein the image comprises at least one pixel;
wherein determining the first reflectance value comprises using a pixel value of the at least one pixel.

8. The method of claim 5 , wherein the first object is a first human face, the estimated distance is a first estimated distance, the method further comprising:
determining a second object in the scene;
determining that the second object is a second human face;
determining a second reflectance value for the second human face;
calculating a second estimated distance between the imaging device and the second human face;
determining an average of the first estimated distance and the second estimated distance; and
positioning a lens of the imaging device based at least in part on the average.

9. The method of claim 5 , wherein identifying the first object comprises comparing the first reflected illuminance value to a set of reflectance values stored in a reflectance table; and
determining an orientation of the first object, wherein the orientation is associated with the first reflected illuminance value in the reflectance table.

10. The method of claim 5 , wherein the reflected illuminance value is a first reflected illuminance value and the estimated distance is a first estimated distance, the method further comprising:
illuminating the scene at the first illuminance value after determining the distance;
determining a second reflected illuminance value;
determining a second reflectance value for the first object;
determining a second estimated distance between the imaging device and the first object;
determining that the second estimated distance is different than the first estimated distance; and
adjusting a focus setting of the imaging device based on the second estimated distance.

11. The method of claim 10 , wherein adjusting the focus setting of the imaging device based on the second estimated distance comprises directing sound towards the first object and adjusting audio intensity based at least in part on the second estimated distance.

12. The method of claim 11 , wherein the first object is a person, the method further comprising:
determining that the p

Assignees

• Amazon Tech Inc

Inventors

• Solh Mashhour
• Poon Tsz Lok

Classifications

• G06T7/50 Primary

  Depth or shape recovery · CPC title

• G06V10/75

  Organisation of the matching processes, e.g. simultaneous or sequential comparisons of image or video features; Coarse-fine approaches, e.g. multi-scale approaches; using context analysis; Selection of dictionaries · CPC title

• H04N23/611

  where the recognised objects include parts of the human body · CPC title

• H04N23/74

  by influencing the scene brightness using illuminating means · CPC title

• G06F18/22

  Matching criteria, e.g. proximity measures · CPC title