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.

Sign up for GitHub

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

Jump to bottom

Using Caffe for RGB-D dataset with depth map output #955

Closed
ghost opened this issue Aug 20, 2014 · 5 comments
Closed

Using Caffe for RGB-D dataset with depth map output #955

ghost opened this issue Aug 20, 2014 · 5 comments
Labels
question

Comments

@ghost
Copy link

ghost commented Aug 20, 2014

Hello.
I have a dataset composed of images (3 x 256 x 256) and corresponding depth maps (256 x 256).
Would it be possible use Caffe for training this dataset (input as images and output as depth maps)?
I found that Caffe can deal with multi label data (#523), but this multi-label can deal with only 3 multiple labels.
I appreicate your help!
@shelhamer
Copy link
Member

The short answer is yes. For details please ask questions on the caffe-users mailing list. As of the latest release we prefer to keep issues reserved for Caffe development. Thanks!

@shelhamer shelhamer changed the title Using Caffe for RGB-D dataset Using Caffe for RGB-D dataset with depth map output Aug 21, 2014
shelhamer added a commit to shelhamer/caffe that referenced this issue Sep 3, 2014
@shelhamer
Merge pull request BVLC#955 from kloudkl/data-layers
f4727dc 
Further simplify the data layers and extend the MemoryDataLayer.

- design BaseDataLayer and BasePrefectingDataLayer for simplification
- refactor DataLayer, ImageDataLayer, and WindowDataLayer
- pull up `transform_param` into layer message for de-duplication
- add transformation to MemoryDataLayer
mitmul pushed a commit to mitmul/caffe that referenced this issue Sep 30, 2014
@shelhamer
Merge pull request BVLC#955 from kloudkl/data-layers
d7f3a1e 
Further simplify the data layers and extend the MemoryDataLayer.

- design BaseDataLayer and BasePrefectingDataLayer for simplification
- refactor DataLayer, ImageDataLayer, and WindowDataLayer
- pull up `transform_param` into layer message for de-duplication
- add transformation to MemoryDataLayer
RazvanRanca pushed a commit to RazvanRanca/caffe that referenced this issue Nov 4, 2014
@shelhamer
Merge pull request BVLC#955 from kloudkl/data-layers
e762b83 
Further simplify the data layers and extend the MemoryDataLayer.

- design BaseDataLayer and BasePrefectingDataLayer for simplification
- refactor DataLayer, ImageDataLayer, and WindowDataLayer
- pull up `transform_param` into layer message for de-duplication
- add transformation to MemoryDataLayer
@demondan
Copy link

hello,how to deal with this question ? can you tell me?

@ghost
Copy link
Author

ghost commented Mar 14, 2015

Hello @demondan. This was my previous project. Unfortunately I could not figure out this. I am sorry!

@shelhamer
Copy link
Member

See #1698 for how to prepare data for matrix (like an image) prediction and ground truth.

@Codermanok
Copy link

Could you share you image depth dataset? or how can I get the image depth map dataset?Thank you.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
question
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
3 participants
@shelhamer @Codermanok @demondan