some upsample methods in compute

参考：

1. Jeremy Jordan--An overview of semantic image segmentation
2. stanford cs231 Fei-Fei Li & Justin Johnson & Serena YeungLecture 11 -May 10, 20171Lecture 11:Detection and Segmentation
3. https://github.com/vdumoulin/conv_arithmetic

There are a few different approaches that we can use to upsample the resolution of a feature map. Whereas pooling operations downsample the resolution by summarizing a local area with a single value (ie. average or max pooling), "unpooling" operations upsample the resolution by distributing a single value into a higher resolution.

image.png

However, transpose convolutions are by far the most popular approach as they allow for us to develop a learned upsampling.

image.png

Whereas a typical convolution operation will take the dot product of the values currently in the filter's view and produce a single value for the corresponding output position, a transpose convolution essentially does the opposite. For a transpose convolution, we take a single value from the low-resolution feature map and multiply all of the weights in our filter by this value, projecting those weighted values into the output feature map.

A simplified 1D example of upsampling through a transpose operation.png