Ask Your Question

Revision history [back]

click to hide/show revision 1
initial version

answered 2018-04-27 13:58:27 -0600

berak gravatar image

this is not a real answer, more a collection of hints, but it' ll be quite long, so no fit for a comment.

i somehow think, you're underestimating the number of "moving parts", such an application will need.

  • believe it or not, you need a webserver, and hosting. whatever framework you use for routing (django, flask, bottle, plain wsgi, even) is the least relevant part of your pipeline (imho. yea, webdevs will cry out loud, but i'm not one.)
  • you probably want your users to access their local webcam (and that from a webbrowser), to take images, any camera attached to your box is useless (did you think of it ?)
  • so that will require javascript (served from your server) to access it, do a face-detection, and send an image home for classification (something like this might do a good job here)
  • then, you want to process the images on your server, do face classification and such,
  • and send back results to the browser


here's some (fictional) routing layout:

/               -- the main, index page
/js/js1.js
/js/js2.js      -- javascript needed to access the webcam, detect & crop an image, send back image
/upload         -- receive the image
/results        -- the face classification result
/results/img    -- if you sent a thumbnail back or such

this is not a real answer, more a collection of hints, but it' ll be quite long, so no fit for a comment.

i somehow think, you're underestimating the number of "moving parts", such an application will need.

  • believe it or not, you need a webserver, and hosting. whatever framework you use for routing (django, flask, bottle, plain wsgi, even) is the least relevant part of your pipeline (imho. yea, webdevs will cry out loud, but i'm not one.)
  • you probably want your users to access their local webcam (and that from a webbrowser), to take images, any camera attached to your box is useless (did you think of it ?)
  • so that will require javascript (served from your server) to access it, do a face-detection, and send an image home for classification (something like this might do a good job here)
  • then, you want to process the images on your server, do face classification and such,
  • and send back results to the browser


here's some (fictional) routing layout:

/               -- the main, index page
/js/js1.js
/js/js2.js      -- javascript needed to access the webcam, detect & crop an image, send back image
img
/upload         -- receive the image
/results        -- the face classification result
/results/img    -- if you sent a thumbnail back or such

this is not a real answer, more a collection of hints, but it' ll be quite long, so no fit for a comment.

i somehow think, you're underestimating the number of "moving parts", such an application will need.

  • believe it or not, you need a webserver, and hosting. whatever framework you use for routing (django, flask, bottle, plain wsgi, even) is the least relevant part of your pipeline (imho. yea, webdevs will cry out loud, but i'm not one.)
  • you probably want your users to access their local webcam (and that from a webbrowser), to take images, any camera attached to your box is useless (did you think of it ?)
  • so that will require javascript (served from your server) to access it, do a face-detection, and send an image home for classification (something like this might do a good job here)
  • then, you want to process the images on your server, do face classification and such,
  • and send back results to the browser


here's some (fictional) routing layout:

/               -- the main, index page
/js/js1.js
/js/js2.js      -- javascript needed to access the webcam, detect & crop an image, a face, send back img
/upload         -- receive the image
/results        -- the face classification result
/results/img    -- if you sent a thumbnail back or such

this is not a real answer, more a collection of hints, but it' ll be quite long, so no fit for a comment.

i somehow think, you're underestimating the number of "moving parts", such an application will need.

  • believe it or not, you need a webserver, and hosting. whatever framework you use for routing (django, flask, bottle, plain wsgi, even) is the least relevant part of your pipeline (imho. yea, webdevs will cry out loud, but i'm not one.)
  • you probably want your users to access their local webcam (and that from a webbrowser), to take images, any camera attached to your box is useless (did you think of it ?)
  • so that will require javascript (served from your server) to access it, do a face-detection, face-detection (again inside the webbrowser), and send an image home for classification (something like this might do a good job here)
  • then, you want to process the images on your server, do face classification and such,
  • and send back results to the browser


here's some (fictional) routing layout:

/               -- the main, index page
/js/js1.js
/js/js2.js      -- javascript needed to access the webcam, detect & crop a face, send back img
/upload         -- receive the image
/results        -- the face classification result
/results/img    -- if you sent a thumbnail back or such

this is not a real answer, more a collection of hints, but it' ll be quite long, so no fit for a comment.

i somehow think, you're underestimating the number of "moving parts", such an application will need.

  • believe it or not, you need a webserver, and hosting. whatever framework you use for routing (django, flask, bottle, plain wsgi, even) is the least relevant part of your pipeline (imho. yea, webdevs will cry out loud, but i'm not one.)
  • you probably want your users to access their local webcam (and that from a webbrowser), to take images, any camera attached to your box is useless (did you think of it ?)
  • so that will require javascript (served from your server) to access it, do a face-detection (again inside the webbrowser), and send an a (cropped) image home for classification (something like this might do a good job here)
  • then, you want to process the images on your server, do face classification and such,
  • and send back results to the browser


here's some (fictional) routing layout:

/               -- the main, index page
/js/js1.js
/js/js2.js      -- javascript needed to access the webcam, detect & crop a face, send back img
/upload         -- receive the image
/results        -- the face classification result
/results/img    -- if you sent a thumbnail back or such

this is not a real answer, more a collection of hints, but it' ll be quite long, so no fit for a comment.

i somehow think, you're underestimating the number of "moving parts", such an application will need.

  • believe it or not, you need a webserver, and hosting. whatever framework you use for routing (django, flask, bottle, plain wsgi, even) is the least relevant part of your pipeline (imho. yea, webdevs will cry out loud, but i'm not one.)
  • you probably want your users to access their local webcam (and that from a webbrowser), to take images, any camera attached to your box is useless (did you think of it ?)
  • so that will require javascript (served from your server) to access it, do a face-detection (again inside the webbrowser), and send a (cropped) image home for classification (something like this might do a good job here)
  • then, you want to process the images on your server, do face classification and such,
  • and send back results to the browser


here's some (fictional) routing layout:

/               -- (get) the main, index page
/js/js1.js
/js/js2.js      -- (get) javascript needed to access the webcam, detect & crop a face, send back img
/upload         -- (post) receive the image
/results        -- (get) the face classification result
/results/img    -- (get) if you sent a thumbnail back or such

this is not a real answer, more a collection of hints, but it' ll be quite long, so no fit for a comment.

i somehow think, you're underestimating the number of "moving parts", such an application will need.

  • believe it or not, you need a webserver, and hosting. whatever framework you use for routing (django, flask, bottle, plain wsgi, even) is the least relevant part of your pipeline (imho. yea, webdevs will cry out loud, but i'm not one.)
  • you probably want your users to access their local webcam (and that from a webbrowser), to take images, any camera attached to your box is useless (did you think of it ?)
  • so that will require javascript (served from your server) to access it, do a face-detection (again inside the webbrowser), and send a (cropped) image home for classification (something like this might do a good job here)
  • then, you want to process the images on your server, do face classification and such,
  • and send back results to the browser


here's some (fictional) routing layout:

/               -- (get) the main, index page
/js/js1.js
/js/js2.js      -- (get) javascript needed to access the webcam, detect & crop a face,  send back img
/upload         -- (post) receive the image
/results        -- (get) the face classification result
/results/img    -- (get) if you sent a thumbnail back or such

this is not a real answer, more a collection of hints, but it' ll be quite long, so no fit for a comment.

i somehow think, you're underestimating the number of "moving parts", such an application will need.need, so heere's some "food for thought":

  • believe it or not, you need a webserver, and hosting. whatever framework you use for routing (django, flask, bottle, plain wsgi, even) is the least relevant part of your pipeline (imho. yea, webdevs will cry out loud, but i'm not one.)
  • you probably want your users to access their local webcam (and that from a webbrowser), to take images, any camera attached to your box is useless (did you think of it ?)
  • so that will require javascript (served from your server) to access it, do a face-detection (again inside the webbrowser), and send a (cropped) image home for classification (something like this might do a good job here)
  • then, you want to process the images on your server, do face classification and such,
  • and send back results to the browser


here's some (fictional) routing layout:

/               -- (get) the main, index page
/js/js1.js
/js/js2.js      -- (get) javascript needed to access the webcam, detect & crop a face, 
                         send back img
/upload         -- (post) receive the image
/results        -- (get) the face classification result
/results/img    -- (get) if you sent a thumbnail back or such

this is not a real answer, more a collection of hints, but it' ll be quite long, so no fit for a comment.

i somehow think, you're underestimating the number of "moving parts", such an application will need, so heere's here's some "food for thought":

  • believe it or not, you need a webserver, and hosting. whatever framework you use for routing (django, flask, bottle, plain wsgi, even) is the least relevant part of your pipeline (imho. yea, webdevs will cry out loud, but i'm not one.)
  • you probably want your users to access their local webcam (and that from a webbrowser), to take images, any camera attached to your box is useless (did you think of it ?)
  • so that will require javascript (served from your server) to access it, do a face-detection (again inside the webbrowser), and send a (cropped) image home for classification (something like this might do a good job here)
  • then, you want to process the images on your server, do face classification and such,
  • and send back results to the browser


here's some (fictional) routing layout:

/               -- (get) the main, index page
/js/js1.js
/js/js2.js      -- (get) javascript needed to access the webcam, detect & crop a face, 
                         send back img
/upload         -- (post) receive the image
/results        -- (get) the face classification result
/results/img    -- (get) if you sent a thumbnail back or such

this is not a real answer, more a collection of hints, but it' ll be quite long, so no fit for a comment.

i somehow think, you're underestimating the number of "moving parts", such an application will need, so here's some "food for thought":

  • believe it or not, you need a webserver, and hosting. whatever framework you use for routing (django, flask, bottle, plain wsgi, even) is the least relevant part of your pipeline (imho. yea, webdevs will cry out loud, but i'm not one.)
  • you probably want your users to access their local webcam (and that from a webbrowser), to take images, any camera attached to your box is useless (did you think of it ?)
  • so that will require javascript (served from your server) to access it, do a face-detection (again inside the webbrowser), and send a (cropped) image home for classification (something like this might do a good job here)
  • then, you want to process the images on your server, do face classification and such,
  • and send back results to the browser


here's some (fictional) routing layout:layout (you'll have to implement a function to handle each of those):

/               -- (get) the main, index page
/js/js1.js
/js/js2.js      -- (get) javascript needed to access the webcam, detect & crop a face, 
                         send back img
/upload         -- (post) receive the image
/results        -- (get) the face classification result
/results/img    -- (get) if you sent a thumbnail back or such

this is not a real answer, more a collection of hints, but it' ll be quite long, so no fit for a comment.

i somehow think, you're underestimating the number of "moving parts", such an application will need, so here's some "food for thought":

  • believe it or not, you need a webserver, and hosting. whatever framework you use for routing (django, flask, bottle, plain wsgi, even) is the least relevant part of your pipeline (imho. yea, webdevs will cry out loud, but i'm not one.)
  • you probably want your users to access their local webcam (and that from a webbrowser), to take images, any camera attached to your box is useless (did you think of it ?)
  • so that will require javascript (served from your server) to access it, do a face-detection (again inside the webbrowser), and send a (cropped) image home for classification (something like this might do a good job here)
  • then, you want to process the images on your server, do face classification and such,
  • and send back results to the browser


here's some (fictional) routing layout (you'll have to implement a function to handle each of those):

/               -- (get) the main, index page
/js/js1.js
/js/js2.js      -- (get) javascript needed to access the webcam, detect & crop a face, 
                         send back img
/upload         -- (post) receive the image
/results        -- (get) the face classification result
result (html)
/results/img    -- (get) if you sent a thumbnail back or such
Copyright OpenCV foundation, 2012-2018. Content on this site is licensed under a Creative Commons Attribution Share Alike 3.0 license.
about | faq | help | privacy policy | terms of service
Powered by Askbot version 0.10.2