Profile

Cover photo
Davis King
Worked at Shell TechWorks
Attended Ohio State University
Lives in Boston, MA
141 followers|1,167,131 views
AboutPostsPhotosYouTube

Stream

Davis King

Shared publicly  - 
 
Easily Create High Quality Object Detectors with Deep Learning
A few years ago I added an implementation of the max-margin object-detection algorithm  (MMOD) to dlib. This tool has since become quite popular as it frees the user from tedious tasks like hard negative mining.  You simply label things in images and it lea...
A few years ago I added an implementation of the max-margin object-detection algorithm (MMOD) to dlib. This tool has since become quite popular as it frees the user from tedious tasks like hard negative mining. You simply la...
2
Davis King's profile photoRishabh Agrawal's profile photo吴岸城's profile photo
6 comments
 
hi davis: i want use dlib detect in camera with , how to speed up ?
Add a comment...

Davis King

Shared publicly  - 
 
Dlib 19.1 Released
cuDNN 5.1 is out and it isn't completely backwards compatible with cuDNN 5.0 due to a bug in cuDNN 5.1.  For the curious, in cuDNN 5.1  cudnnGetConvolutionBackwardFil terAlgorithm() will select the winograd algorithm even when the conv descriptor has a stri...
cuDNN 5.1 is out and it isn't completely backwards compatible with cuDNN 5.0 due to a bug in cuDNN 5.1. For the curious, in cuDNN 5.1 cudnnGetConvolutionBackwardFilterAlgorithm() will select the winograd algorithm even when ...
2
Add a comment...
 
 
A Clean C++11 Deep Learning API
Dlib 19.0 is out and it has a lot of new features, like new elastic net and quadratic program solvers. But the feature I'm most excited about is the new deep learning API. There are a lot of existing deep learning frameworks, but none of them have clean C++...
1 comment on original post
1
Add a comment...

Davis King

Shared publicly  - 
 
A Clean C++11 Deep Learning API
Dlib 19.0 is out and it has a lot of new features, like new elastic net and quadratic program solvers. But the feature I'm most excited about is the new deep learning API. There are a lot of existing deep learning frameworks, but none of them have clean C++...
4
13
Philip Barkes's profile photo

Davis King

Shared publicly  - 
 
Reinforcement Learning, Control, and 3D Visualization
Over the last few months I've spent a lot of time studying optimal control and reinforcement learning. Aside from reading, one of the best ways to learn about something is to do it yourself, which in this case means a lot of playing around with the well kno...
3
2
Devesh Parekh's profile photoudaya chandran's profile photoDavis King's profile photo
5 comments
 
Sometimes that's just what happens.  You will need to train your own detector with improved accuracy to fix it.  Look at the training example program to see how this can be done.
Add a comment...

Davis King

Shared publicly  - 
 
Dlib 18.6 released: Make your own object detector!
I just posted the next version of dlib, v18.6.  There are a bunch of nice changes, but the most exciting addition is a tool for creating histogram-of-oriented-gradient (HOG) based object detectors.  This is a technique for detecting semi-rigid objects in im...
I just posted the next version of dlib, v18.6.  There are a bunch of nice changes, but the most exciting addition is a tool for creating histogram-of-oriented-gradient (HOG) based object detectors.  This is a technique for de...
3
Mykhaylo Sorochev's profile photo
 
Looks cool! Could you help me to solve one object tracking task ? mike.sorochev@gmail.com
Add a comment...

Davis King

Shared publicly  - 
 
Hipsterize Your Dog With Deep Learning
I'm getting ready to make the next dlib release, which should be out in a few days, and I thought I would point out a humorous new example program.  The dog hipsterizer ! It uses dlib's new deep learning tools to detect dogs looking at the camera. Then it u...
I'm getting ready to make the next dlib release, which should be out in a few days, and I thought I would point out a humorous new example program. The dog hipsterizer! It uses dlib's new deep learning tools to detect dogs...
4
Richard Sanderson's profile photoDavis King's profile photo
4 comments
 
Ha, me too. Supposedly my parents are looking for a photo. Best I can do right now is one of their current dogs all tarted up. Looks like bigfoot though :)
https://plus.google.com/photos/...
Add a comment...

Davis King

Libraries & Frameworks  - 
 
 
A Clean C++11 Deep Learning API
Dlib 19.0 is out and it has a lot of new features, like new elastic net and quadratic program solvers. But the feature I'm most excited about is the new deep learning API. There are a lot of existing deep learning frameworks, but none of them have clean C++...
1 comment on original post
2
Add a comment...

Davis King

Discussion  - 
 
 
A Clean C++11 Deep Learning API
Dlib 19.0 is out and it has a lot of new features, like new elastic net and quadratic program solvers. But the feature I'm most excited about is the new deep learning API. There are a lot of existing deep learning frameworks, but none of them have clean C++...
1 comment on original post
14
9
Add a comment...

Davis King

AI Software  - 
 
 
Reinforcement Learning, Control, and 3D Visualization
Over the last few months I've spent a lot of time studying optimal control and reinforcement learning. Aside from reading, one of the best ways to learn about something is to do it yourself, which in this case means a lot of playing around with the well kno...
5 comments on original post
6
1
Add a comment...

Davis King

Shared publicly  - 
3
Yatong Zhang's profile photo
 
COOL! Does this demo have some code to share?
Add a comment...

Davis King

AI Software  - 
 
Check this out if you are interested in tracking objects in videos
30
11
Iron Bug's profile photoJay Raley's profile photoMatthias Mueller's profile photoDavis King's profile photo
7 comments
 
I wrote my implementation before their Matlab code was released so I didn't do a comparison.  I think I created what they described in the paper, but I did notice that the accuracy wasn't as high as what they reported.  I assumed it's because they include additional features in their code like re-acquisition logic which weren't mentioned in the paper, but since the code wasn't available I couldn't check.  Or maybe there is just some flaw in my implementation.  I can't say.
Add a comment...
Work
Occupation
Software Engineer
Employment
  • Shell TechWorks
    Sr Software Engineer, 2014 - 2016
  • MIT Lincoln Laboratory
    Associate Staff, 2011 - 2014
  • Northrop Grumman
    Principal Engineer, 2005 - 2011
  • Intel
    2003 - 2004
  • svlsResearch
    2016 - present
Places
Map of the places this user has livedMap of the places this user has livedMap of the places this user has lived
Currently
Boston, MA
Links
YouTube
Contributor to
Education
  • Ohio State University
  • Johns Hopkins
Basic Information
Gender
Male