FaceCompressor Class¶
Namespace: id3.Face
Definition¶
Properties¶
Name |
Type |
Description |
|---|---|---|
Model used for face compression decoding. |
||
Model used for face compression encoding. |
||
Processing unit where to run the compression and decompression processes. |
||
int |
Number of threads to be used for inference. |
Methods¶
Name |
Description |
|---|---|
Detects the largest face in an image, aligns and crops it into a Bioseal portrait (256x192), and compresses it into a fixed-size 500-byte template using a Deep Learning encoder model. |
|
Compresses a pre-aligned 4:3 Bioseal portrait image (256x192) into a fixed-size 500-byte compressed template using a Deep Learning encoder model. |
|
Aligns and crops an already detected face into a Bioseal portrait (256x192), and compresses it into a fixed-size 500-byte template using a Deep Learning encoder model. |
|
Detects the largest face in an image, crops it using a tighter, eye-distance-based face placement (different from the Bioseal portrait used by |
|
Compresses a pre-cropped portrait image to WebP so that its size does not exceed the given byte budget. |
|
Crops an already detected face using a tighter, eye-distance-based face placement (different from the Bioseal portrait used by |
|
Decompresses a 500-byte compressed template back into a 4:3 portrait image (256x192) using a Deep Learning decoder model. |
|
Decompresses a WebP buffer produced by |