-
Notifications
You must be signed in to change notification settings - Fork 57
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
- Loading branch information
Showing
2 changed files
with
18 additions
and
10 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -1,26 +1,32 @@ | ||
# captioning | ||
### From Captions to Visual Concepts and Back ### | ||
Code for detecting visual concepts in images. | ||
|
||
#### Installation Instructions #### | ||
0. Create directory, checkout code, caffe, coco-hooks | ||
|
||
```shell | ||
git clone git@github.com:s-gupta/im2cap.git code | ||
git clone git@github.com:pdollar/coco.git coco | ||
``` | ||
|
||
0. Make caffe and pycaffe | ||
```shell | ||
git clone git@github.com:s-gupta/caffe.git caffe | ||
cd caffe | ||
git checkout mil | ||
make -j 16 | ||
make pycaffe | ||
cd | ||
``` | ||
### Get the data ### | ||
0. Get the COCO images | ||
|
||
0. Get the caffe image net models | ||
|
||
0. Get the pre-trained models | ||
0. Get the COCO images, caffe imagenet models, pretrained models on COCO. | ||
``` shell | ||
# Get the COCO images, splits, ground truth | ||
wget http://www.cs.berkeley.edu/~sgupta/captions/data/data.tgz && tar -xf data.tgz | ||
# Get the caffe imagenet models | ||
wget http://www.cs.berkeley.edu/~sgupta/captions/data/caffe-data.tgz && tar -xf caffe-data.tgz | ||
# Get the pretrained models | ||
wget http://www.cs.berkeley.edu/~sgupta/captions/data/pretrained-coco.tgz && tar -xf pretrained-coco.tgz | ||
### Testing the model ### | ||
#### Training, Testing the model #### | ||
```cd code``` and execute relevant commands from the file scripts/scripts_all.py |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,2 @@ | ||
#### Directory for storing the prototxt files, snapshots and evaluations #### | ||
0. vgg - Contains the model that was trained on COCO for the CVPR 15 paper. |