Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Add a simple C++ wrapper for ACF Detector #5

Open
wants to merge 5 commits into
base: master
Choose a base branch
from
Open

Add a simple C++ wrapper for ACF Detector #5

wants to merge 5 commits into from

Conversation

zeyuanxy
Copy link

Hi Piotr, ACF is really fantastic! I have read your paper about that and I am impressed by its great performance and speed, thus I add some other code and one READMEs for a simple C++ wrapper for the function acfDetect. Thus it will be easy to deploy as the heavy-weighted MATLAB is not required to be installed. It is really a honor for me to help with this, Thanks!

@pdollar
Copy link
Owner

pdollar commented May 22, 2015

Hi Edison --

Thanks, this sounds great! I won't have time in the immediate future to go
through the code and merge it into the toolbox, but I'm sure people will
find it useful regardless :-)

-Piotr

On Fri, May 22, 2015 at 2:17 AM, Edison notifications@github.com wrote:

Hi Piotr, ACF is really fantastic! I have read your paper about that and I
am impressed by its great performance and speed, thus I add some other code
and one READMEs for a simple C++ wrapper for the function acfDetect. Thus
it will be easy to deploy as the heavy-weighted MATLAB is not required to

be installed. It is really a honor for me to help with this, Thanks!

You can view, comment on, or merge this pull request online at:

#5
Commit Summary

  • add new ACF detect function and MATLAB dll file
  • change files location
  • add cpp_wrapper for acfDetector
  • remove unused files
  • add README for cpp_wrappers

File Changes

Patch Links:


Reply to this email directly or view it on GitHub
#5.

@mammadjv
Copy link

@zeyuanxy Hello .. sorry I have a question .. the compilation of code is in Matlab runtime?? We want to run out code with gcc compiler in DSP board . is there any way for doing that??

@nkhdiscovery
Copy link

nkhdiscovery commented Sep 27, 2016

@mohammadjv6
Hi Mohammad Javad,
First, you may need to talk about such issues in "issues", not pull request. I'm sorry to answer your question here, too.
By the way, what you asked for is almost impossible in first place because there is no open-source implementation for ACF in C/C++, at least I have not found any and I have not implemented my own yet. After all, if there was a nice implementation which you could port on your DSP module, it would not guarantee run-time performance as you may need re-implement most of your code in ASM decelerations with knowledge of your DSP assembly language and optimization on parts which can be optimized with your DSP chip. Texas Instruments DSPs have a nice compiler for porting all or parts of your C/C++ code to run on DSP module entirely, but it usually causes lots of overheads by default that you may reach even lower performances. The best is to find/write a good C/C++ implementation - having in mind that you're going to use a DSP in future, so functions should use some declaratives like I don't know, #ifdef DSP_ACCL by your own - and then break the code to parts which can be accelerated using a DSP and implement DSP compatible parts. If you're using TI DSPs, you can first port a function to DSP and measure the performance, and re-implement it with assembly if needed. You may need to take a look at http://www.ti.com/lit/wp/spry175/spry175.pdf first, then you can implement a better general purpose code with OpenCV.

Good luck.

@CamilleMaurice
Copy link

Hello!
I'd like to use the wrapper, but I could not find how I can download the new files. Thank you
Camille

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

None yet

5 participants