Commit Graph

81 Commits

Author SHA1 Message Date
montag 47afcbf35a Tls cert bugfix (#262)
* Bugfix for systems that cannot generate openssl certs. Includes self-signed certs that expire in 10 years. Removed automatic cert generation from install-deps.sh. Updated create-cert.sh to manually generate 10year certs (or any custom duration).

* Docker config changes to support web demo TLS
2017-05-01 13:40:54 -04:00
montag 37ba2ac982 TLS migration for demos/web (#260)
What does this PR do?

Minimally invasive migration to TLS for the http (8000) and websocket (9000) endpoints for the real time web demo.

Where should the reviewer start?

Install the demo like a normal install. During the install-deps.sh script it will prompt for questions to generate a local self-signed cert. Anything can be entered into the cert. Start the demo as normal but connect on https://domain:http_port instead of http.

How should this PR be tested?

The first step is the cert generation (added to install-deps.sh for convenience). The second step is ensuring the two endpoints are brought up. The next step is loading the web page and accepting the self-signed cert. Reloading the web app everything should now be secure, e.g. no errors or warnings and the video works as normal.

Any background context you want to provide?

I tried to be as minimal as possible so the changes tends to follow the existing structure rather than a refactor of anything major. With that said, SimpleHTTPServer doesn't work with TLS via the -m flag so that is now a short script. I updated all the html/js files to point to https/wss. I also updated the js for Firefox's change from navigator.mozGetUserMedia to navigator.mediaDevices.getUserMedia.

What are the relevant issues?

#75

Questions:

Do the docs need to be updated?
Yes. I updated the script's docs in demos/web but didn't make any changes outside of demos/web

Does this PR add new (Python) dependencies?
I don't think so.
2017-04-28 06:05:12 -04:00
Andrew 71ee6ee747 Make OpenFace work with Python 3 (#231)
* Make OpenFace Python 3 compatible, retaining Python 2 compatibility.
Tests could be parameterised to test for either Python 2 or 3.
Tests currently fail due to different library calculation outcomes?

* Tiny change to remove the dependency on nosetest3.

* Dockerfile updates for OpenFace + Python3 + Dlib19.2 + OpenCV3.2

* Revert "Dockerfile updates for OpenFace + Python3 + Dlib19.2 + OpenCV3.2"

This reverts commit ec08fc9226.

* Change for python2 backwards compatibility

* Change for python2 backwards compatibility

* Change to ensure python2 backwards compatibility

* Change to ensure python2 backwards compatibility
2017-04-24 17:50:16 -04:00
Brandon Amos 96b45fd0eb
Use https for the model downloads. 2017-02-07 13:43:42 -05:00
Brandon Amos 320cf3d728
Update links to our models. 2017-02-07 11:56:49 -05:00
NMA 041c90f0c1 Added --multi option to infer operation to show a list of faces detected (#189)
* Added --multi option to infer operation to show a list of faces detected in image

* Added testing for infer --multi demo
2016-09-22 06:51:04 -04:00
Brandon Amos 7f6413fca9 Remove unused code. 2016-09-15 08:43:43 -04:00
Brandon Amos 6b453b0b00 sphere demo: Add link to model download. 2016-09-15 08:01:26 -04:00
Brandon Amos 0b2f07127d cls-3d -> sphere 2016-09-11 12:27:04 -04:00
Brandon Amos 1745b19dfa
cls-3d: Fix pep warnings. 2016-09-02 14:27:42 -04:00
Brandon Amos 9c5580d7d1
cls-3d: Bump alpha up. Add text. 2016-09-02 13:33:59 -04:00
Brandon Amos f5c71ad47b
Initial commit with the rest of cls-3d.py 2016-09-02 08:36:22 -04:00
Gabriele Farina da93d45236
Initial commit of projectS and projectC to cls-3d.py 2016-09-01 23:27:30 -04:00
Rasmus Haglund 611d05af47 Updated to work with new openface api (#177)
* Updated to work with new openface api

* Prevent the save from breaking when getRep returns None
2016-08-06 21:21:33 -04:00
Gu Wang 80014e0e97 Update classifier.py (#176)
* Update classifier.py

Add an option to train with grid search svm.

* Update classifier.py
2016-08-06 21:19:21 -04:00
Vijayenthiran Subramaniam fa2a3a5d20 Unknown people benchmarking with lfw dataset. (#171)
* Added few classifiers to demos/classifier.py; demos/classifier_test_train.py - for test train accuracy estimation; demos/classifier_webcam.py for webcam face recognition

* Two stage classifier in webcam demo

* Renamed classifier_test_train; cleaned up demos/classifier_webcam.py

* Initial commit of working unknown evaluation. Clean up pending

* Clean up lfw-classification-unknown.py and fix flake8 warnings
2016-07-28 14:38:14 -04:00
Brandon Amos 84576d252d Remove align_v2.
align_v2 manually computes the same affine transformation as align_v1
2016-07-15 15:14:58 -04:00
Brandon Amos 026288afa6 Merge pull request #155 from heman4t/master
Updated demo/web/vendor deps
2016-06-28 12:40:12 -04:00
hemantsangwan efa3ab8007 Updated web/vendor deps 2016-06-28 10:47:47 +05:30
Dante Knowles 3e679b9fb9 Added align dlib v2 2016-06-07 20:26:28 +00:00
Brandon Amos 3b9b99c3b2 Bump scikit-learn requirements for #137. 2016-05-13 18:24:17 -04:00
Brandon Amos 98c42a057c Add LDA option to the classifier. 2016-04-03 15:36:58 -04:00
Brandon Amos 2035eeb8a5 cls->clf 2016-03-18 16:19:08 -04:00
Brandon Amos 437603e45d Classification Demo: Fix typo and formatting. 2016-03-18 16:15:58 -04:00
Brandon Amos 3a5ce4e70a Add GMM option to the classification demo. 2016-03-18 15:26:40 -04:00
Brandon Amos cca19b4c6a Classification Demo: Output where classifier is saved to. 2016-03-18 14:53:33 -04:00
Brandon Amos d4ee384d5e demos/vis-outputs: Add aligned example images, fix image.load usage, and output where the images are being saved to
Thanks @pppoe! https://github.com/cmusatyalab/openface/pull/107
2016-03-09 22:26:12 -05:00
Brandon Amos 18575f7199 Web demo: bower: Update metadata. 2016-03-09 18:56:56 -05:00
Brandon Amos c79cc4a73e Classification demo: Use a linear SVM with C=1 instead of grid search. 2016-02-19 14:18:02 -05:00
Brandon Amos cfe3a83e2f websocket server: Use store_true for --cuda. 2016-02-05 20:08:18 -05:00
Brandon Amos 19d9d8c703 Add more details to start-servers.sh
Thanks @justinsteele
2016-01-20 23:18:17 -05:00
Brandon Amos 9d1bd99579 Update copyright notices.
sed -i "/[Cc]opyright.*2015.*Carnegie Mellon University/s/2015/2015-2016/;"  $(git grep '[Cc]opyright.*Carnegie Mellon University'  | sed 's/:.*//' | uniq)
2016-01-13 14:52:52 -05:00
Brandon Amos f96b206765 Use nn4.small2.v1 by default. 2016-01-12 15:46:57 -05:00
Brandon Amos ff06eff893 Docker: Move Python dependencies to Ubuntu package binaries. 2016-01-11 16:12:45 -05:00
Brandon Amos 1de080d65b Web Demo: Use outer eyes and nose. 2016-01-11 10:31:38 -05:00
Brandon Amos 434dc1ca9f Fix flake8 warnings. 2016-01-07 18:43:25 -05:00
Brandon Amos 6246fada83 Update code to nn4.v2 for #55. 2016-01-07 18:28:11 -05:00
Brandon Amos 4357fdccae Update websocket-server to latest API. 2015-12-31 13:02:36 -05:00
Brandon Amos e15d11fbd7 Finish API changes and inline Python docs for #73. 2015-12-30 13:14:30 -05:00
Brandon Amos 1f7b50efa2 Refactor Python interface. 2015-12-29 19:57:29 -05:00
Brandon Amos dceb6a910b flake8 improvements. 2015-12-24 11:45:00 -05:00
Brandon Amos e1a8ccbe37 Classifier Demo: Take multiple images on input. 2015-12-24 11:06:52 -05:00
Brandon Amos 3e5be95a3c Web demo: Assume the HTTP and WebSocket servers are the same so the web demo works in Docker without any modifications. 2015-12-23 18:42:40 -05:00
Brandon Amos 122f70ce3b Remove unused skimage dependency. 2015-12-22 22:14:35 -05:00
Brandon Amos fbde622805 Demos: Remove sys.path appends and unused dlib imports. 2015-12-22 21:33:27 -05:00
Brandon Amos ec6aa45d89 System-wide install of dlib and OpenFace.
Resolves #11
2015-12-22 20:58:06 -05:00
Brandon Amos 027f9dcf8e Real-Time Web Demo: Simplify server initialization. 2015-12-19 18:17:27 -05:00
Brandon Amos 7d73946836 Web demo: server.py -> websocket-server.py 2015-12-19 17:46:57 -05:00
Brandon Amos aec7de78dc Explictly set txaio.use_twisted.
https://groups.google.com/forum/#!topic/cmu-openface/zbWEuYCf7_w
2015-12-18 10:52:36 -05:00
Brandon Amos 918d3af141 Add port option to server. 2015-12-18 10:52:36 -05:00