dfdx / facedatasets.jl

Simple(r) access to face-related datasets

FaceDatasets

Package for making access to popular face datasets easier.

General usage

load_images(<dataset_name>, <dataset_params>...)  # faces
load_shapes(<dataset_name>, <dataset_params>...)  # face landmarks
load_labels(<dataset_name>, <dataset_params>...)  # provided labels

For example, to load CootesDataset, enter:

load_images(CootesDataset)

Note, that all methods return iterable objects: for small datasets they are just arrays, for larger iterators are returned instead. You can always materialize them using:

collect(load_images(...))

Available datasets

Cootes images

CootesDataset contains images from Tim Cootes' work on active appearance models. These images come prepacked, so you can use them for testing.

Supported functions:

• load_images(CootesDataset)
• load_shapes(CootesDataset)

Cohn-Kanade+ dataset

CKDataset contains images from Cohn-Kanade+ Expression Database. To install this dataset, download it from this page and unpack into a directory of your choice. Example of expected directory layout:

$ tree -L 2
.
├── cohn-kanade-images
│   ├── S005
│   ├── S010
│   ├── S011
│   ...
│   └── S999
├── Emotion
│   ├── S005
│   ├── S010
│   ├── S011
│   ...
│   └── S999
└── Landmarks
    ├── S005
    ├── S010
    ├── S011
    ...
    └── S999

Supported functions:

• load_images(CKDataset, datadir, opts...)
• load_shapes(CKDataset, datadir, opts...)
• load_labels(CKDataset, datadir, opts...)

where datadir is base dir for CK dataset and labels are numbers representing 6 basic emotions + neutral facial expression.

Options:

• start - image index to start with
• count - number of images to return
• indexes - concrete indexes to return (start and count are ignored)
• resizeratio - resize image by this value

Cohn-Kanade+ (max only) dataset

CKMaxDataset - same as Cohn-Kanade+ dataset, but contains only images with maximally expressed emotion (~500 images). Only resizeration option is supported, though.

TODO: KaggleFERDataset TODO: PutFrontalDataset

facedatasets.jl's Issues

ERROR: Gray not defined

After running the sample script: "multi.jl" in the ActiveAppearanceModel package, I get a few warnings and errors:

Warning: could not import Base.Text into Tk
WARNING: Color is deprecated, please use Colors instead. See https://github.com/JuliaGraphics/Colors.jl
ERROR: Gray not defined

From the following file:

ActiveAppearanceModels/examples/multi.jl, in expression starting on line 4

Which is referenced from:

FaceDatasets/src/FaceDatasets.jl, in expression starting on line 6

Which is referenced from:

FaceDatasets/src/core.jl, in expression starting on line 5

Which contains the following line:

Base.convert{T}(::Type{Matrix{Float64}}, img::Image{Gray{T}}) = convert(Array{Float64, 2}, convert(Array, img))

I tried modifying the dependency on the deprecated package Color to Colors but that didn't seem to fix the error.

Any idea what the issue might be?

Info about upcoming removal of packages in the General registry

As described in https://discourse.julialang.org/t/ann-plans-for-removing-packages-that-do-not-yet-support-1-0-from-the-general-registry/ we are planning on removing packages that do not support 1.0 from the General registry. This package has been detected to not support 1.0 and is thus slated to be removed. The removal of packages from the registry will happen approximately a month after this issue is open.

To transition to the new Pkg system using Project.toml, see https://github.com/JuliaRegistries/Registrator.jl#transitioning-from-require-to-projecttoml.
To then tag a new version of the package, see https://github.com/JuliaRegistries/Registrator.jl#via-the-github-app.

If you believe this package has erroneously been detected as not supporting 1.0 or have any other questions, don't hesitate to discuss it here or in the thread linked at the top of this post.