AWE Toolbox is intended for automatic evaluation of feature extractor methods on image datasets. AWE Dataset is already included, as well as some extraction methods.
You can use (and modify, but leave headers in the code) the toolbox and the dataset for non-commercial use. In your paper please cite:
Ž. Emeršič, V. Štruc, and P. Peer: »Ear Recognition: More than a Survey«,
Neurocomputing, 2017
Download bibtex here.
Annotated Web Ears Dataset – AWE Dataset | Bibliographic information
| Authors: | Žiga Emeršič, Vitomir Štruc, Peter Peer |
|---|
| Item URL: | ears.fri.uni-lj.si/about#awed |
|---|
| Year of publication: | 2017 |
|---|
| Type: | corpus (COBISS type 2.20) |
|---|
| Size: | 1000 images, 100 identities; 30.5 MB (30.0 MiB) |
|---|
| Publisher: | Faculty of Computer and Information Science, University of Ljubljana |
|---|
| Description: | The database contains 1.000 images of 100 persons. Images were taken from the wild (acquired from the internet) and contain the following annotations: gender, ethnicity, accessories, occlusions, head pitch, head roll, head yaw, head side, central tragus point – all stored in JSON files. |
|---|
Annotated Web Ears Toolbox – AWE Toolbox | Bibliographic information
| Authors: | Žiga Emeršič, Vitomir Štruc, Peter Peer |
|---|
| Item URL: | ears.fri.uni-lj.si/about#awet |
|---|
| Year of publication: | 2017 |
|---|
| Type: | software (COBISS type 2.21) |
|---|
| Size: | 10.2 MB (9.8 MiB) |
|---|
| Publisher: | Faculty of Computer and Information Science, University of Ljubljana |
|---|
| Description: | The toolbox is written in Matlab. You need Mex for it to run. The version here does not include the database due to bibliographical reasons, you need to download it separately. The toolbox should work out of the box if you have environment set correctly. If it does not work, see the help section. |
|---|