Intro

  • Reference : ICML2016

    • There are two networks in this architecture. One is the Generator and the other is the Discriminator.

    Discrminator learns from real data distribution to give scores on images generated from Generator.

    Generator uses the scores given by Discrminator to update its parameters.

    Atrribute vectors are sent into these two networks in order to generate images based on given conditions.

    Details

    There are 40 human attributes listed below as input conditions.

    For each image,its paired conditions are in the form of a vector of 40 length. Each value in the vector corresponds to a attribute,and is marked 1 if the person in that image has that attribute,otherwise is marked 0.

    The attribute vector of 40 length is concatenated to

    a noise vector of 100 length.

    ---Do linear scaling to a larger size.

    ---Reshape it to let it has spatial dimension.

    ---Go throug 4 layers of deconvolution,pooling,Relu.

    ---The final output has a size of [256,256,3],

    which is the generated image.

    The input images first go through 4

    layers of convolution,pooling,Relu.

    ---Copy all the values in the attribute

    vector to let it has a spatial dimension.

    ---Concatenate these two tensors.

    ---Reshape and linear-scale to size 1,

    which represents its score

    Finally, modify its original structure.

    Identity Block from Residual Net is added before each deconv layer in discriminator.

    Results

    Below shows the generated images from the generator during training phase.

    Below shows the generated images and their corresponding attributes (unseen distribution during training).

    Elements

    Text

    This is bold and this is strong. This is italic and this is emphasized. This is superscript text and this is subscript text. This is underlined and this is code: for (;;) { ... }. Finally, this is a link.

    Heading Level 2

    Heading Level 3

    Heading Level 4

    Heading Level 5
    Heading Level 6

    Blockquote

    Fringilla nisl. Donec accumsan interdum nisi, quis tincidunt felis sagittis eget tempus euismod. Vestibulum ante ipsum primis in faucibus vestibulum. Blandit adipiscing eu felis iaculis volutpat ac adipiscing accumsan faucibus. Vestibulum ante ipsum primis in faucibus lorem ipsum dolor sit amet nullam adipiscing eu felis.

    Preformatted

    i = 0;

while (!deck.isInOrder()) {
    print 'Iteration ' + i;
    deck.shuffle();
    i++;
}

print 'It took ' + i + ' iterations to sort the deck.';

    Lists

    Unordered

    • Dolor pulvinar etiam.
    • Sagittis adipiscing.
    • Felis enim feugiat.

    Alternate

    • Dolor pulvinar etiam.
    • Sagittis adipiscing.
    • Felis enim feugiat.

    Ordered

    1. Dolor pulvinar etiam.
    2. Etiam vel felis viverra.
    3. Felis enim feugiat.
    4. Dolor pulvinar etiam.
    5. Etiam vel felis lorem.
    6. Felis enim et feugiat.

    Icons

    Actions

    Table

    Default

    Name Description Price
    Item One Ante turpis integer aliquet porttitor. 29.99
    Item Two Vis ac commodo adipiscing arcu aliquet. 19.99
    Item Three Morbi faucibus arcu accumsan lorem. 29.99
    Item Four Vitae integer tempus condimentum. 19.99
    Item Five Ante turpis integer aliquet porttitor. 29.99
    100.00

    Alternate

    Name Description Price
    Item One Ante turpis integer aliquet porttitor. 29.99
    Item Two Vis ac commodo adipiscing arcu aliquet. 19.99
    Item Three Morbi faucibus arcu accumsan lorem. 29.99
    Item Four Vitae integer tempus condimentum. 19.99
    Item Five Ante turpis integer aliquet porttitor. 29.99
    100.00

    Buttons

    • Disabled
    • Disabled

    Form