github/deepface
1
1
Fork 0
mirror of https://github.com/serengil/deepface.git synced 2025-06-07 03:55:21 +00:00
Code Issues Packages Projects Releases Wiki Activity

some links

Browse Source
This commit is contained in:
Sefik Ilkin Serengil 2020-05-22 23:55:56 +03:00 committed by GitHub
parent 8efef29b19
commit 91a4b01991
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 2 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
README.md
View File

@ -39,7 +39,7 @@ dataset = [
resp_obj = DeepFace.verify(dataset)
```
Items of resp_obj might be unsorted when you pass multiple instances to verify function.
Items of resp_obj might be unsorted when you pass multiple instances to verify function. Please check the item indexes in the response object.
## Face recognition models
@ -72,7 +72,7 @@ DeepFace.verify("img1.jpg", "img2.jpg", model_name = "VGG-Face", model = model)
## Similarity
These models actually find the vector embeddings of faces. In other words, we use face recognition models as [`autoencoders`](https://sefiks.com/2018/03/23/convolutional-autoencoder-clustering-images-with-neural-networks/). Decision of verification is based on the distance between vectors. Distance could be found by different metrics such as [`Cosine Similarity`](https://sefiks.com/2018/08/13/cosine-similarity-in-machine-learning/), Euclidean Distance and L2 form. The default configuration finds the **cosine similarity**. You can alternatively set the similarity metric while verification as demostratred below.
Face recognition models are regular [`convolutional neural networks`](https://sefiks.com/2018/03/23/convolutional-autoencoder-clustering-images-with-neural-networks/) and they are responsible to represent face photos as vectors. Decision of verification is based on the distance between vectors. We can classify pairs if its distance is less than a [`threshold`](https://sefiks.com/2020/05/22/fine-tuning-the-threshold-in-face-recognition/). Distance could be found by different metrics such as [`Cosine Similarity`](https://sefiks.com/2018/08/13/cosine-similarity-in-machine-learning/), Euclidean Distance and L2 form. The default configuration finds the **cosine similarity**. You can alternatively set the similarity metric while verification as demostratred below.
```python
result = DeepFace.verify("img1.jpg", "img2.jpg", model_name = "VGG-Face", distance_metric = "cosine")