Update demos with results from the new models.

This commit is contained in:
Brandon Amos 2015-12-10 21:05:52 -05:00
parent 4a94b51314
commit 12a8648959
3 changed files with 13 additions and 13 deletions

View File

@ -112,4 +112,4 @@ def getRep(imgPath):
for (img1, img2) in itertools.combinations(args.imgs, 2):
d = getRep(img1) - getRep(img2)
print("Comparing {} with {}.".format(img1, img2))
print(" + Squared l2 distance between representations: {}".format(np.dot(d, d)))
print(" + Squared l2 distance between representations: {:0.3f}".format(np.dot(d, d)))

View File

@ -13,18 +13,18 @@ Eric Clapton were generated with
|---|---|---|---|
| <img src='https://raw.githubusercontent.com/cmusatyalab/openface/master/images/examples/lennon-1.jpg' width='200px'></img> | <img src='https://raw.githubusercontent.com/cmusatyalab/openface/master/images/examples/lennon-2.jpg' width='200px'></img> | <img src='https://raw.githubusercontent.com/cmusatyalab/openface/master/images/examples/clapton-1.jpg' width='200px'></img> | <img src='https://raw.githubusercontent.com/cmusatyalab/openface/master/images/examples/clapton-2.jpg' width='200px'></img> |
The following table shows that a distance threshold of `0.5` would
The following table shows that a distance threshold of `0.8` would
distinguish these two people.
In practice, further experimentation should be done on the distance threshold.
On our LFW experiments, the mean threshold across multiple
experiments is 0.71 &plusmn; 0.027,
experiments is about 0.68,
see [accuracies.txt](https://github.com/cmusatyalab/openface/blob/master/evaluation/lfw.nn4.v1.epoch-177/accuracies.txt).
| Image 1 | Image 2 | Distance |
|---|---|---|
| Lennon 1 | Lennon 2 | 0.310 |
| Lennon 1 | Clapton 1 | 1.241 |
| Lennon 1 | Clapton 2 | 1.056 |
| Lennon 2 | Clapton 1 | 1.386 |
| Lennon 2 | Clapton 2 | 1.073 |
| Clapton 1 | Clapton 2 | 0.259 |
| Lennon 1 | Lennon 2 | 0.298 |
| Lennon 1 | Clapton 1 | 0.985 |
| Lennon 1 | Clapton 2 | 0.855 |
| Lennon 2 | Clapton 1 | 0.904 |
| Lennon 2 | Clapton 2 | 0.955 |
| Clapton 1 | Clapton 2 | 0.712 |

View File

@ -88,7 +88,7 @@ Run the classifier on your images with:
| Person | Image | Prediction | Confidence |
|---|---|---|---|
| Carell | <img src='https://raw.githubusercontent.com/cmusatyalab/openface/master/images/examples/carell.jpg' width='200px'></img> | SteveCarell | 0.78 |
| Adams | <img src='https://raw.githubusercontent.com/cmusatyalab/openface/master/images/examples/adams.jpg' width='200px'></img> | AmyAdams | 0.87 |
| Lennon 1 (Unknown) | <img src='https://raw.githubusercontent.com/cmusatyalab/openface/master/images/examples/lennon-1.jpg' width='200px'></img> | DavidBoreanaz | 0.28 |
| Lennon 2 (Unknown) | <img src='https://raw.githubusercontent.com/cmusatyalab/openface/master/images/examples/lennon-2.jpg' width='200px'></img> | DavidBoreanaz | 0.56 |
| Carell | <img src='https://raw.githubusercontent.com/cmusatyalab/openface/master/images/examples/carell.jpg' width='200px'></img> | SteveCarell | 0.96 |
| Adams | <img src='https://raw.githubusercontent.com/cmusatyalab/openface/master/images/examples/adams.jpg' width='200px'></img> | AmyAdams | 0.98 |
| Lennon 1 (Unknown) | <img src='https://raw.githubusercontent.com/cmusatyalab/openface/master/images/examples/lennon-1.jpg' width='200px'></img> | DavidBoreanaz | 0.27 |
| Lennon 2 (Unknown) | <img src='https://raw.githubusercontent.com/cmusatyalab/openface/master/images/examples/lennon-2.jpg' width='200px'></img> | DavidBoreanaz | 0.43 |