�ݺ�ߣshows by User: aurot

�ݺ�ߣshows by User: aurot / http://www.slideshare.net/images/logo.gif �ݺ�ߣshows by User: aurot / Tue, 23 May 2017 12:59:45 GMT �ݺ�ߣShare feed for �ݺ�ߣshows by User: aurot Auro tripathy - Localizing with CNNs /slideshow/auro-tripathy-localizing-with-cnns/76253036 aurotripathylocalisingwithcnnsforposting-170523125945
Locating objects in images (“detection”) quickly and efficiently enables object tracking and counting applications on embedded visual sensors (fixed and mobile). By 2012, progress on techniques for detecting objects in images – a topic of perennial interest in computer vision – had plateaued, and techniques based on histogram of oriented gradients (HOG) were state of the art. Soon, though, convolutional neural networks (CNNs), in addition to classifying objects, were also beginning to become effective at simultaneously detecting objects. Research in CNN-based object detection was jump-started by the groundbreaking region-based CNN (R-CNN). We’ll follow the evolution of neural network algorithms for object detection, starting with R-CNN and proceeding to Fast R-CNN, Faster R-CNN, “You Only Look Once” (YOLO), and up to the latest Single Shot Multibox detector. In this talk, we’ll examine the successive innovations in performance and accuracy embodied in these algorithms – which is a good way to understand the insights behind effective neural-network-based object localization. We’ll also contrast bounding-box approaches with pixel-level segmentation approaches and present pros and cons.]]>
Locating objects in images (“detection”) quickly and efficiently enables object tracking and counting applications on embedded visual sensors (fixed and mobile). By 2012, progress on techniques for detecting objects in images – a topic of perennial interest in computer vision – had plateaued, and techniques based on histogram of oriented gradients (HOG) were state of the art. Soon, though, convolutional neural networks (CNNs), in addition to classifying objects, were also beginning to become effective at simultaneously detecting objects. Research in CNN-based object detection was jump-started by the groundbreaking region-based CNN (R-CNN). We’ll follow the evolution of neural network algorithms for object detection, starting with R-CNN and proceeding to Fast R-CNN, Faster R-CNN, “You Only Look Once” (YOLO), and up to the latest Single Shot Multibox detector. In this talk, we’ll examine the successive innovations in performance and accuracy embodied in these algorithms – which is a good way to understand the insights behind effective neural-network-based object localization. We’ll also contrast bounding-box approaches with pixel-level segmentation approaches and present pros and cons.]]> Tue, 23 May 2017 12:59:45 GMT /slideshow/auro-tripathy-localizing-with-cnns/76253036 aurot@slideshare.net(aurot) Auro tripathy - Localizing with CNNs aurot Locating objects in images (“detection”) quickly and efficiently enables object tracking and counting applications on embedded visual sensors (fixed and mobile). By 2012, progress on techniques for detecting objects in images – a topic of perennial interest in computer vision – had plateaued, and techniques based on histogram of oriented gradients (HOG) were state of the art. Soon, though, convolutional neural networks (CNNs), in addition to classifying objects, were also beginning to become effective at simultaneously detecting objects. Research in CNN-based object detection was jump-started by the groundbreaking region-based CNN (R-CNN). We’ll follow the evolution of neural network algorithms for object detection, starting with R-CNN and proceeding to Fast R-CNN, Faster R-CNN, “You Only Look Once” (YOLO), and up to the latest Single Shot Multibox detector. In this talk, we’ll examine the successive innovations in performance and accuracy embodied in these algorithms – which is a good way to understand the insights behind effective neural-network-based object localization. We’ll also contrast bounding-box approaches with pixel-level segmentation approaches and present pros and cons. <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/aurotripathylocalisingwithcnnsforposting-170523125945-thumbnail.jpg?width=120&height=120&fit=bounds" /> Locating objects in images (“detection”) quickly and efficiently enables object tracking and counting applications on embedded visual sensors (fixed and mobile). By 2012, progress on techniques for detecting objects in images – a topic of perennial interest in computer vision – had plateaued, and techniques based on histogram of oriented gradients (HOG) were state of the art. Soon, though, convolutional neural networks (CNNs), in addition to classifying objects, were also beginning to become effective at simultaneously detecting objects. Research in CNN-based object detection was jump-started by the groundbreaking region-based CNN (R-CNN). We’ll follow the evolution of neural network algorithms for object detection, starting with R-CNN and proceeding to Fast R-CNN, Faster R-CNN, “You Only Look Once” (YOLO), and up to the latest Single Shot Multibox detector. In this talk, we’ll examine the successive innovations in performance and accuracy embodied in these algorithms – which is a good way to understand the insights behind effective neural-network-based object localization. We’ll also contrast bounding-box approaches with pixel-level segmentation approaches and present pros and cons.

Auro tripathy - Localizing with CNNs from Auro Tripathy

]]> 2071 4 https://cdn.slidesharecdn.com/ss_thumbnails/aurotripathylocalisingwithcnnsforposting-170523125945-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post

http://activitystrea.ms/schema/1.0/posted

0 GoogLeNet Insights /slideshow/googlenet-insights/66227863 googlenet-insights-160920190051
Five Insights from GoogLeNet You Could Use In Your Own Deep Learning Nets]]>
Five Insights from GoogLeNet You Could Use In Your Own Deep Learning Nets]]> Tue, 20 Sep 2016 19:00:51 GMT /slideshow/googlenet-insights/66227863 aurot@slideshare.net(aurot) GoogLeNet Insights aurot Five Insights from �GoogLeNet �You Could Use In Your Own Deep Learning Nets <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/googlenet-insights-160920190051-thumbnail.jpg?width=120&height=120&fit=bounds" /> Five Insights from �GoogLeNet �You Could Use In Your Own Deep Learning Nets

GoogLeNet Insights from Auro Tripathy

]]> 6011 21 https://cdn.slidesharecdn.com/ss_thumbnails/googlenet-insights-160920190051-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post

http://activitystrea.ms/schema/1.0/posted

0 Back-propagation Primer /slideshow/backpropagation-primer/57543167 back-propagation-primer-160127065811
Welcome to a primer on the back-propagation (of errors) as it applies to the training of neural networks. We answer the question, what's the contribution of the back-propagation-technique?]]>
Welcome to a primer on the back-propagation (of errors) as it applies to the training of neural networks. We answer the question, what's the contribution of the back-propagation-technique?]]> Wed, 27 Jan 2016 06:58:11 GMT /slideshow/backpropagation-primer/57543167 aurot@slideshare.net(aurot) Back-propagation Primer aurot Welcome to a primer on the back-propagation (of errors) as it applies to the training of neural networks. We answer the question, what's the contribution of the back-propagation-technique? <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/back-propagation-primer-160127065811-thumbnail.jpg?width=120&height=120&fit=bounds" /> Welcome to a primer on the back-propagation (of errors) as it applies to the training of neural networks. We answer the question, what's the contribution of the back-propagation-technique?

Back-propagation Primer from Auro Tripathy

]]> 1129 9 https://cdn.slidesharecdn.com/ss_thumbnails/back-propagation-primer-160127065811-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post

http://activitystrea.ms/schema/1.0/posted

0 Of knights-and-drawbridges-nat-behaviour /aurot/of-knightsanddrawbridgesnatbehaviour of-knights-and-drawbridges-nat-behaviour-150526010326-lva1-app6892
Understanding NAT with a simple analogy.]]>
Understanding NAT with a simple analogy.]]> Tue, 26 May 2015 01:03:26 GMT /aurot/of-knightsanddrawbridgesnatbehaviour aurot@slideshare.net(aurot) Of knights-and-drawbridges-nat-behaviour aurot Understanding NAT with a simple analogy. <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/of-knights-and-drawbridges-nat-behaviour-150526010326-lva1-app6892-thumbnail.jpg?width=120&height=120&fit=bounds" /> Understanding NAT with a simple analogy.

Of knights-and-drawbridges-nat-behaviour from Auro Tripathy

]]> 669 4 https://cdn.slidesharecdn.com/ss_thumbnails/of-knights-and-drawbridges-nat-behaviour-150526010326-lva1-app6892-thumbnail.jpg?width=120&height=120&fit=bounds presentation 000000 http://activitystrea.ms/schema/1.0/post

http://activitystrea.ms/schema/1.0/posted

0 A Random Forest Approach To Skin Detection With R /slideshow/a-random-forest-approach-to-skin-detection-with-r/15371898 arandomforestapproachtoskindetectionwithr-13540352993178-phpapp01-121127105541-phpapp01
A Random Forest Approach To Skin Detection With R]]>
A Random Forest Approach To Skin Detection With R]]> Tue, 27 Nov 2012 10:55:41 GMT /slideshow/a-random-forest-approach-to-skin-detection-with-r/15371898 aurot@slideshare.net(aurot) A Random Forest Approach To Skin Detection With R aurot A Random Forest Approach To Skin Detection With R <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/arandomforestapproachtoskindetectionwithr-13540352993178-phpapp01-121127105541-phpapp01-thumbnail.jpg?width=120&height=120&fit=bounds" /> A Random Forest Approach To Skin Detection With R

A Random Forest Approach To Skin Detection With R from Auro Tripathy

]]> 1539 3 https://cdn.slidesharecdn.com/ss_thumbnails/arandomforestapproachtoskindetectionwithr-13540352993178-phpapp01-121127105541-phpapp01-thumbnail.jpg?width=120&height=120&fit=bounds presentation 000000 http://activitystrea.ms/schema/1.0/post

http://activitystrea.ms/schema/1.0/posted

0 Latent Semanctic Analysis Auro Tripathy /slideshow/latent-semanctic-analysis-auro-tripathy/13156437 latentsemancticanalysisseptember72009-13385218146362-phpapp02-120531224009-phpapp02
LSA Overview - Covers Singular Value Decomposition, Dimensionality Reduction, LSA in Information Retrieval]]>
LSA Overview - Covers Singular Value Decomposition, Dimensionality Reduction, LSA in Information Retrieval]]> Thu, 31 May 2012 22:39:39 GMT /slideshow/latent-semanctic-analysis-auro-tripathy/13156437 aurot@slideshare.net(aurot) Latent Semanctic Analysis Auro Tripathy aurot LSA Overview - Covers Singular Value Decomposition, Dimensionality Reduction, LSA in Information Retrieval <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/latentsemancticanalysisseptember72009-13385218146362-phpapp02-120531224009-phpapp02-thumbnail.jpg?width=120&height=120&fit=bounds" /> LSA Overview - Covers Singular Value Decomposition, Dimensionality Reduction, LSA in Information Retrieval

Latent Semanctic Analysis Auro Tripathy from Auro Tripathy

]]> 1342 6 https://cdn.slidesharecdn.com/ss_thumbnails/latentsemancticanalysisseptember72009-13385218146362-phpapp02-120531224009-phpapp02-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post

http://activitystrea.ms/schema/1.0/posted

0 HTTP Live Streaming /slideshow/http-live-streaming-10069443/10069443 httplivestreaming-1320749605449-phpapp01-111108045820-phpapp01
Overview of Apple\'s HTTP Live Streaming (HLS)]]>
Overview of Apple\'s HTTP Live Streaming (HLS)]]> Tue, 08 Nov 2011 04:56:24 GMT /slideshow/http-live-streaming-10069443/10069443 aurot@slideshare.net(aurot) HTTP Live Streaming aurot Overview of Apple\'s HTTP Live Streaming (HLS) <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/httplivestreaming-1320749605449-phpapp01-111108045820-phpapp01-thumbnail.jpg?width=120&height=120&fit=bounds" /> Overview of Apple\'s HTTP Live Streaming (HLS)

HTTP Live Streaming from Auro Tripathy

]]> 20299 9 https://cdn.slidesharecdn.com/ss_thumbnails/httplivestreaming-1320749605449-phpapp01-111108045820-phpapp01-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post

http://activitystrea.ms/schema/1.0/posted

0 https://cdn.slidesharecdn.com/profile-photo-aurot-48x48.jpg?cb=1591244878 • Founder, ShatterLine Labs, a deep learning computer vision startup in stealth. • Software dev mgr for $1B/year revenue product-line (Cisco cable box, HP connected TV). • Apps engineering mgr for seven years for category-defining products (Intel, CCube); $40M rev. • Product mgr for three years, dev tools; scrum product owner, one year, device connectivity. • Software developer and product architect for seven years, video & image processing systems. • Creative solver; four patents granted. http://www.shatterline.com https://cdn.slidesharecdn.com/ss_thumbnails/aurotripathylocalisingwithcnnsforposting-170523125945-thumbnail.jpg?width=320&height=320&fit=bounds slideshow/auro-tripathy-localizing-with-cnns/76253036 Auro tripathy - Local... https://cdn.slidesharecdn.com/ss_thumbnails/googlenet-insights-160920190051-thumbnail.jpg?width=320&height=320&fit=bounds slideshow/googlenet-insights/66227863 GoogLeNet Insights https://cdn.slidesharecdn.com/ss_thumbnails/back-propagation-primer-160127065811-thumbnail.jpg?width=320&height=320&fit=bounds slideshow/backpropagation-primer/57543167 Back-propagation Primer