Skip to content

heckedbyfurki/Face-Recognition-SDK-Android

Β 
Β 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

Β 

History

19 Commits
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 

Repository files navigation

✨ Update: Our API portal is now live, offering free APIs for various AI solutions, including face recognition, liveness detection, and ID document recognition. Make sure to check it out at https://getapi.faceonlive.com and take advantage of our free offerings.

Mask-Aware Face Recognition SDK For Android

NIST FRVT Certified, Fast, Accurate, Mask-Aware Face Recognition SDK with Liveness Detection

www.faceonlive.com  www.faceonlive.com  www.faceonlive.com  www.faceonlive.com

πŸ’ͺ Partnership / Cooperation / Project Discussion

 

πŸŽ‰ Try It Yourself

Get it on Google Play

pro1.mp4

πŸ‘ Supporters

Stargazers repo roster for @faceonlive/Face-Recognition-SDK-Android Forkers repo roster for @faceonlive/Face-Recognition-SDK-Android

πŸƒ Usage SDK

1. Setup Environment

  • Create New Project with Android Studio.

  • Place SDK library ttvface.aar into app / libs.

  • Include SDK library to app / build.gradle file.

dependencies {
    ...
    implementation files('libs/ttvface.aar')
    ...
}
  • Import library in code and call functions.
import com.ttv.face.FaceEngine;

...
// Set License (To obtain license, please contact us)
FaceEngine.getInstance(this).setActivation(license);

// Init with Max Detectable Face Count
FaceEngine.getInstance(this).init(8);

// Detect Face
List<FaceResult> faceInfoList = new ArrayList<>();
faceInfoList = FaceEngine.getInstance(this).detectFace(bitmap);

2. Error Code

F_OK = 0,
F_LICENSE_KEY_ERROR = -1,
F_LICENSE_APPID_ERROR = -2,
F_LICENSE_EXPIRED = -3,
F_INIT_ERROR = -4,

3. Classes

- FaceResult

Type Name Description
Rect rect Face rectangle coordinates
int liveness Liveness status: 0 for spoof, 1 for real, less than 0 for unknown
int gender Gender classification result
int mask Mask presence: 0 for no mask, 1 for mask
int age Age estimation result
float yaw Yaw angle: -45 to 45 degrees
float roll Roll angle: -45 to 45 degrees
float pitch Pitch angle: -45 to 45 degrees
byte[] feature 2056-byte facial feature data
byte[] faceData Encrypted facial data
int orient Face orientation: 1 for no rotation, 2 for 90Β° rotation, 3 for 270Β° rotation, 4 for 180Β° rotation
int faceId Face ID in the tracking face mode
public class FaceResult {
    public Rect rect;
    public int liveness;
    public int gender;
    public int mask;
    public int age;
    public float yaw;
    public float roll;
    public float pitch;
    public byte[] feature;
    public byte[] faceData;
    public int    orient;
    public int faceId;
    
    public FaceResult() {
    }
}

4. API Introduction

- Set Activation

int setActivation(String license)
Name Description
Function Activate SDK with license input
Parameters 0: license String
Return Value F_OK: Succeed, Failed otherwise
import com.ttv.face.FaceEngine;
...
int activated = FaceEngine.getInstance(this).setActivation(license);

- Detect faces from bitmap object

List<FaceResult> detectFace(Bitmap bitmap)
Name Description
Function Get Detection Result from Bitmap image
Parameters Android Bitmap format Image
Return Value List of FaceResult
import com.ttv.face.FaceEngine;
...
List<FaceResult> detectResult = FaceEngine.getInstance(this).detectFace(bitmap);

- Detect faces from yuv data

List<FaceResult> detectFace(byte[] nv21, int width, int height)
Name Description
Function Get Detection Result from Bitmap image
Parameters NV21 YUV ByteArray with it's width and height
Return Value List of FaceResult
import com.ttv.face.FaceEngine;
...
List<FaceResult> detectResult = FaceEngine.getInstance(this).detectFace(nv21, width, height);

- Extract feature with the face detection result

int extractFeature(Bitmap bitmap, boolean isRegister, List<FaceResult> faceResults)
Name Description
Function Extract Face Feature from Detected Face
Parameters Android Bitmap format Image, extract feature for register or not
Return Value 0: Succeed, -1: Failed
If successful, the feature should be get the feature in FaceResult
import com.ttv.face.FaceEngine;
...
List<FaceResult> detectResult = FaceEngine.getInstance(this).extractFeature(bitmap, isRegister, faceResults);

- Extract feature with the face detection result

int extractFeature(byte[] nv21, int width, int height, boolean isRegister, List faceResults)
Name Description
Function Extract Face Feature from Detected Face
Parameters NV21 YUV ByteArray with Width and Height, extract feature for register or not
Return Value 0: Succeed, -1: Failed
If successful, the feature should be get the feature in FaceResult
import com.ttv.face.FaceEngine;
...
List<FaceResult> detectResult = FaceEngine.getInstance(this).extractFeature(bitmap, isRegister, faceResults);

- Compare similarity between two faces

float compareFace(byte[] feat1, byte[] feat2)
Name Description
Function Check two faces are same or not from Feat Array
Parameters Face Features to compare
Return Value Score of Face Comparison (Threshold for same faces 0.82)
import com.ttv.face.FaceEngine;
...
float score = FaceEngine.getInstance(this).compareFace(feat1, feat2);

- Check liveness from bitmap

int livenessProcess(Bitmap bitmap, List<FaceResult> faceResults)
Name Description
Function Check Liveness from Detected Face
Parameters Android Bitmap format Image
Return Value 0: Succeed, -1: Failed
If successful, set Liveness Value to FaceResults
import com.ttv.face.FaceEngine;
...
float score = FaceEngine.getInstance(this).livenessProcess(bitmap, faceResults);

- Check liveness from yuv data

int livenessProcess(byte[] nv21, int width, int height, List<FaceResult> faceResults)
Name Description
Function Check Liveness from Detected Face
Parameters NV21 YUV ByteArray with Width and Height
Return Value 0: Succeed, -1: Failed
If successful, Set Liveness Value to FaceResults
import com.ttv.face.FaceEngine;
...
float score = FaceEngine.getInstance(this).livenessProcess(nv21, width, height, faceResults);

- Get facial attribute(Gender, Age, Yaw, Roll, Pitch, Mask) from bitmap

int faceAttrProcess(Bitmap bitmap, List<FaceResult> faceResults)
Name Description
Function Detect Attributes(Gender, Age, Yaw, Roll, Pitch, Mask) from Face
Parameters Android Bitmap format Image
Return Value 0: Succeed, -1: Failed
If successful, Set Attributes Value to FaceResults
import com.ttv.face.FaceEngine;
...
float score = FaceEngine.getInstance(this).faceAttrProcess(bitmap, faceResults);

- Get facial attribute(Gender, Age, Yaw, Roll, Pitch, Mask) from yuv data

Function: Detect Attributes(Liveness, Gender, Age, Yaw, Roll, Pitch, Mask) from Face
Name Description
Function Detect Attributes(Gender, Age, Yaw, Roll, Pitch, Mask) from Face
Parameters NV21 YUV ByteArray with Width and Height
Return Value 0: Succeed, -1: Failed
If successful, Set Attributes Value to FaceResults
import com.ttv.face.FaceEngine;
...
float score = FaceEngine.getInstance(this).faceAttrProcess(nv21, width, height, faceResults);

- Process all(Extract Feature, Liveness, Gender, Age, Yaw, Roll, Pitch, Mask) from bitmap

int faceAllProcess(Bitmap bitmap, boolean isRegister, List<FaceResult> faceResults)
Name Description
Function Extract Face Feature, Liveness, Get Attributes from Detected Face
Parameters Android Bitmap format Image
Return Value 0: Succeed, -1: Failed
If successful, Set All Attributes Value to FaceResults
import com.ttv.face.FaceEngine;
...
float score = FaceEngine.getInstance(this).faceAllProcess(bitmap, faceResults);

- Process all(Extract Feature, Liveness, Gender, Age, Yaw, Roll, Pitch, Mask) from yuv data

int faceAllProcess(Bitmap bitmap, boolean isRegister, List<FaceResult> faceResults)
Name Description
Function Extract Face Feature, Liveness, Get Attributes from Detected Face
Parameters NV21 YUV ByteArray with Width and Height, extract feature for register or not
Return Value 0: Succeed, -1: Failed
If successful, Set All Attributes Value to FaceResults
import com.ttv.face.FaceEngine;
...
float score = FaceEngine.getInstance(this).faceAllProcess(nv21, width, height, faceResults);

About

Fast, Accurate, Mask-Aware Face Recognition SDK with Liveness Detection

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Java 57.4%
  • Kotlin 42.6%