looking forward to joining the group

[tangsh...@gmail.com]

Hi,

  I am a Chinese undergraduate student who is passionate in  deep learning. 

  1. I use c++ and python since I could program so I have sophisticated experience in both of them.

  2. I start to use caffe about one year ago and kind of familiar with it. (have experience in training, testing model in caffe, be able to add layers in caffe)

  3. Currently, I am developing an emotional analysis system using tensorflow and a detection system which can detect manhole covers on google map street view images in a selected area. The detection part has been finished using faster-rcnn algorithm. The detection system will output a pascal xml files when giving the url of google street view images. The interface is written in python. To finish the part, I have read nearly every codes of faster rcnn so I am familiar with this model.(knowing the adv. and disadv, how to adjust the parameters) 

  In sum, I think I am capable of Improving the framework for CNNs for 2D detection, standardizing all state of the art detection algorithms, python binding, or development of examples and tutorials.

[Marco Antonio]

You could start running the tutorials and trying to come up with a small example that showcases your skills for this idea.

Regards,

Marco.

--
You received this message because you are subscribed to the Google Groups "opendetection" group.
To unsubscribe from this group and stop receiving emails from it, send an email to opendetection+unsubscribe@googlegroups.com.
To post to this group, send email to opende...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/opendetection/83e2f15a-d5f6-4cac-bb0f-151b7bd86019%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[tangsh...@gmail.com]

I have make proposal drafts and submit them. Can you give me some advice about these proposals.

在 2017年3月25日星期六 UTC+8下午12:55:09，Marco Antonio写道：

You could start running the tutorials and trying to come up with a small example that showcases your skills for this idea.

Regards,

Marco.

On Thu, Mar 2, 2017 at 2:50 PM, <tangsh...@gmail.com> wrote:

Hi,

  I am a Chinese undergraduate student who is passionate in  deep learning. 

  1. I use c++ and python since I could program so I have sophisticated experience in both of them.

  2. I start to use caffe about one year ago and kind of familiar with it. (have experience in training, testing model in caffe, be able to add layers in caffe)

  3. Currently, I am developing an emotional analysis system using tensorflow and a detection system which can detect manhole covers on google map street view images in a selected area. The detection part has been finished using faster-rcnn algorithm. The detection system will output a pascal xml files when giving the url of google street view images. The interface is written in python. To finish the part, I have read nearly every codes of faster rcnn so I am familiar with this model.(knowing the adv. and disadv, how to adjust the parameters) 

  In sum, I think I am capable of Improving the framework for CNNs for 2D detection, standardizing all state of the art detection algorithms, python binding, or development of examples and tutorials.

--
You received this message because you are subscribed to the Google Groups "opendetection" group.

To unsubscribe from this group and stop receiving emails from it, send an email to opendetectio...@googlegroups.com.

[tangsh...@gmail.com]

I'm also glad to write some codes about the content in my proposal if you want.

在 2017年3月25日星期六 UTC+8下午12:55:09，Marco Antonio写道：

You could start running the tutorials and trying to come up with a small example that showcases your skills for this idea.

Regards,

Marco.

On Thu, Mar 2, 2017 at 2:50 PM, <tangsh...@gmail.com> wrote:

Hi,

  I am a Chinese undergraduate student who is passionate in  deep learning. 

  1. I use c++ and python since I could program so I have sophisticated experience in both of them.

  2. I start to use caffe about one year ago and kind of familiar with it. (have experience in training, testing model in caffe, be able to add layers in caffe)

  3. Currently, I am developing an emotional analysis system using tensorflow and a detection system which can detect manhole covers on google map street view images in a selected area. The detection part has been finished using faster-rcnn algorithm. The detection system will output a pascal xml files when giving the url of google street view images. The interface is written in python. To finish the part, I have read nearly every codes of faster rcnn so I am familiar with this model.(knowing the adv. and disadv, how to adjust the parameters) 

  In sum, I think I am capable of Improving the framework for CNNs for 2D detection, standardizing all state of the art detection algorithms, python binding, or development of examples and tutorials.

--
You received this message because you are subscribed to the Google Groups "opendetection" group.

To unsubscribe from this group and stop receiving emails from it, send an email to opendetectio...@googlegroups.com.

[shitao tang]

Dear mentor，

I am sorry to say I have other staff to deal with in the summer so I can't do the project. Please don't accept my proposal. My name is Shitao Tang and the email is tangsh...@gmail.com
regards，

Shitao