<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
	<channel>
		<title>Vision on Alok Regmi</title>
		<link>https://alokregmi.com/tags/vision/</link>
		<description>Recent content in Vision on Alok Regmi</description>
		<generator>Hugo</generator>
		<language>en-us</language>
		
		
		
		
			<lastBuildDate>Sat, 02 May 2020 04:04:27 +0545</lastBuildDate>
		
			<atom:link href="https://alokregmi.com/tags/vision/index.xml" rel="self" type="application/rss+xml" />
			<item>
				<title>Efficient Nets</title>
				<link>https://alokregmi.com/braindump/2020/04/efficient-nets/</link>
				<pubDate>Sun, 19 Apr 2020 19:28:53 +0545</pubDate>
				<guid>https://alokregmi.com/braindump/2020/04/efficient-nets/</guid>
				<description>&lt;h2 id=&#34;efficient-nets&#34;&gt;Efficient Nets:&lt;/h2&gt;&#xA;&lt;ol&gt;&#xA;&lt;li&gt;&#xA;&lt;p&gt;Why all this ?&#xA;&amp;ndash;  Scaling CNN’s only in one direction (eg depth only) will result in rapidly&#xA;deteriorating gains relative to the computational increase needed.&lt;/p&gt;&#xA;&lt;p&gt;&amp;ndash; ResNet 1000 isn’t much more accurate than ResNet152 for example,&#xA;as after 100 -150 layer’s gains rapidly drop off.&lt;/p&gt;&#xA;&lt;p&gt;&amp;ndash; Scaling depth, width and resolution all benifits quickly saturate so&#xA;not at all  possibility of sota.&lt;/p&gt;&#xA;&lt;/li&gt;&#xA;&lt;li&gt;&#xA;&lt;p&gt;We now know the problem so let&amp;rsquo;s do this instead:&#xA;&amp;ndash; In order to scale up efficiently, all dimensions of depth,&#xA;width and resolution have to be scaled together,&#xA;and there is an optimal balance for each dimension relative to the others.&#xA;VOILA!!!&lt;/p&gt;&#xA;&lt;p&gt;&amp;ndash;  after an extensive grid search derived the theoretically optimal formula of “compound scaling” using the following co-efficients:&lt;/p&gt;&#xA;&lt;p&gt;Depth = 1.20&lt;/p&gt;&#xA;&lt;p&gt;Width = 1.10&lt;/p&gt;&#xA;&lt;p&gt;Resolution = 1.15&lt;/p&gt;&#xA;&lt;p&gt;Only google can do this kind of computational  search !! :(&lt;/p&gt;&#xA;&lt;/li&gt;&#xA;&lt;li&gt;&#xA;&lt;p&gt;By showing the existing heatmaps in different widths, depths and resolution,&#xA;they show that their model captures more real image like features than some&#xA;nonsense these NNs learn on the way to correct prediction.&#xA;At last!! We can at least have our brothers in image recognition who detects&#xA;images more naturally looking on the inside.&lt;/p&gt;&#xA;&lt;/li&gt;&#xA;&lt;li&gt;&#xA;&lt;p&gt;Now not just large GPUs can do it, my phone can load a model and use it properly.&#xA;Thanks to reduced number of parameters. Thanks to Gooooogogoogogogogogle&amp;rsquo;s computational&#xA;for having this grid search and the paper.&lt;/p&gt;&#xA;&lt;/li&gt;&#xA;&lt;/ol&gt;</description>
			</item>
			<item>
				<title>Convolution</title>
				<link>https://alokregmi.com/braindump/2020/04/convolution/</link>
				<pubDate>Sun, 19 Apr 2020 16:00:43 +0545</pubDate>
				<guid>https://alokregmi.com/braindump/2020/04/convolution/</guid>
				<description>&lt;h2 id=&#34;convolution&#34;&gt;Convolution&lt;/h2&gt;&#xA;&lt;p&gt;Image will be of shape: (H * W * channels)&lt;/p&gt;&#xA;&lt;p&gt;Our kernel will also be of shape (Hk * Wk * channels)&lt;/p&gt;&#xA;&lt;p&gt;The number of channels is same as it is volume convolution.&lt;/p&gt;&#xA;&lt;p&gt;In case of 1 * 1 convolution, the convolution helps decrease or increase the feature map dimensions.&lt;/p&gt;</description>
			</item>
			<item>
				<title>Image Processing Pipeline</title>
				<link>https://alokregmi.com/braindump/2020/04/image-processing-pipeline/</link>
				<pubDate>Sun, 19 Apr 2020 16:00:34 +0545</pubDate>
				<guid>https://alokregmi.com/braindump/2020/04/image-processing-pipeline/</guid>
				<description>&lt;p&gt;Integrating image processing with machine learning consists of following steps:&lt;/p&gt;&#xA;&lt;ol&gt;&#xA;&lt;li&gt;&#xA;&lt;p&gt;Decide your project title/purpose/objectives.&lt;/p&gt;&#xA;&lt;/li&gt;&#xA;&lt;li&gt;&#xA;&lt;p&gt;Collect data, i.e images relevant to your project.&lt;/p&gt;&#xA;&lt;/li&gt;&#xA;&lt;li&gt;&#xA;&lt;p&gt;Preprocess all the collected images. Preprocessing steps may be different for different projects. It basically includes filtering, noise     removal, grayscale conversion, binary image formation, morphological operation, thresholding and so on.&lt;/p&gt;&#xA;&lt;/li&gt;&#xA;&lt;li&gt;&#xA;&lt;p&gt;Divide the preprocessed images into two sets: training and tesing sets.&lt;/p&gt;&#xA;&lt;/li&gt;&#xA;&lt;li&gt;&#xA;&lt;p&gt;Extract features that you think will most probably define your data. (For example, if you are working on a face detection project, one possible feature could be the distance between the two eyes. You can think of a number of such features) (very important step)&lt;/p&gt;&#xA;&lt;/li&gt;&#xA;&lt;/ol&gt;</description>
			</item>
			<item>
				<title>Efficient Nets</title>
				<link>https://alokregmi.com/ai/efficient_nets/</link>
				<pubDate>Fri, 20 Mar 2020 00:00:00 +0000</pubDate>
				<guid>https://alokregmi.com/ai/efficient_nets/</guid>
				<description>&lt;h2 id=&#34;efficient-nets&#34;&gt;Efficient Nets:&lt;/h2&gt;&#xA;&lt;ol&gt;&#xA;&lt;li&gt;&#xA;&lt;p&gt;Why all this ?&#xA;&amp;ndash;  Scaling CNN’s only in one direction (eg depth only) will result in rapidly&#xA;deteriorating gains relative to the computational increase needed.&lt;/p&gt;&#xA;&lt;p&gt;&amp;ndash; ResNet 1000 isn’t much more accurate than ResNet152 for example,&#xA;as after 100 -150 layer’s gains rapidly drop off.&lt;/p&gt;&#xA;&lt;p&gt;&amp;ndash; Scaling depth, width and resolution all benifits quickly saturate so&#xA;not at all  possibility of sota.&lt;/p&gt;&#xA;&lt;/li&gt;&#xA;&lt;li&gt;&#xA;&lt;p&gt;We now know the problem so let&amp;rsquo;s do this instead:&#xA;&amp;ndash; In order to scale up efficiently, all dimensions of depth,&#xA;width and resolution have to be scaled together,&#xA;and there is an optimal balance for each dimension relative to the others.&#xA;VOILA!!!&lt;/p&gt;&#xA;&lt;p&gt;&amp;ndash;  after an extensive grid search derived the theoretically optimal formula of “compound scaling” using the following co-efficients:&lt;/p&gt;&#xA;&lt;p&gt;Depth = 1.20&lt;/p&gt;&#xA;&lt;p&gt;Width = 1.10&lt;/p&gt;&#xA;&lt;p&gt;Resolution = 1.15&lt;/p&gt;&#xA;&lt;p&gt;Only google can do this kind of computational  search !! :(&lt;/p&gt;&#xA;&lt;/li&gt;&#xA;&lt;li&gt;&#xA;&lt;p&gt;By showing the existing heatmaps in different widths, depths and resolution,&#xA;they show that their model captures more real image like features than some&#xA;nonsense these NNs learn on the way to correct prediction.&#xA;At last!! We can at least have our brothers in image recognition who detects&#xA;images more naturally looking on the inside.&lt;/p&gt;&#xA;&lt;/li&gt;&#xA;&lt;li&gt;&#xA;&lt;p&gt;Now not just large GPUs can do it, my phone can load a model and use it properly.&#xA;Thanks to reduced number of parameters. Thanks to Gooooogogoogogogogogle&amp;rsquo;s computational&#xA;for having this grid search and the paper.&lt;/p&gt;&#xA;&lt;/li&gt;&#xA;&lt;/ol&gt;</description>
			</item>
			<item>
				<title>Image Processing Pipeline</title>
				<link>https://alokregmi.com/ai/image_processing_pipeline/</link>
				<pubDate>Fri, 24 Jan 2020 00:00:00 +0000</pubDate>
				<guid>https://alokregmi.com/ai/image_processing_pipeline/</guid>
				<description>&lt;p&gt;Integrating image processing with machine learning consists of following steps:&lt;/p&gt;&#xA;&lt;ol&gt;&#xA;&lt;li&gt;&#xA;&lt;p&gt;Decide your project title/purpose/objectives.&lt;/p&gt;&#xA;&lt;/li&gt;&#xA;&lt;li&gt;&#xA;&lt;p&gt;Collect data, i.e images relevant to your project.&lt;/p&gt;&#xA;&lt;/li&gt;&#xA;&lt;li&gt;&#xA;&lt;p&gt;Preprocess all the collected images. Preprocessing steps may be different for different projects. It basically includes filtering, noise     removal, grayscale conversion, binary image formation, morphological operation, thresholding and so on.&lt;/p&gt;&#xA;&lt;/li&gt;&#xA;&lt;li&gt;&#xA;&lt;p&gt;Divide the preprocessed images into two sets: training and tesing sets.&lt;/p&gt;&#xA;&lt;/li&gt;&#xA;&lt;li&gt;&#xA;&lt;p&gt;Extract features that you think will most probably define your data. (For example, if you are working on a face detection project, one possible feature could be the distance between the two eyes. You can think of a number of such features) (very important step)&lt;/p&gt;&#xA;&lt;/li&gt;&#xA;&lt;/ol&gt;</description>
			</item>
			<item>
				<title>Convolution</title>
				<link>https://alokregmi.com/ai/convolution/</link>
				<pubDate>Thu, 05 Dec 2019 00:00:00 +0000</pubDate>
				<guid>https://alokregmi.com/ai/convolution/</guid>
				<description>&lt;h2 id=&#34;convolution&#34;&gt;Convolution&lt;/h2&gt;&#xA;&lt;p&gt;Image will be of shape: (H * W * channels)&lt;/p&gt;&#xA;&lt;p&gt;Our kernel will also be of shape (Hk * Wk * channels)&lt;/p&gt;&#xA;&lt;p&gt;The number of channels is same as it is volume convolution.&lt;/p&gt;&#xA;&lt;p&gt;In case of 1 * 1 convolution, the convolution helps decrease or increase the feature map dimensions.&lt;/p&gt;</description>
			</item>
	</channel>
</rss>
