# colorization-v2¶

## Use Case and High-Level Description¶

The colorization-v2 model is one of the colorization group of models designed to perform image colorization. Model was trained on ImageNet dataset. For details about this family of models, check out the repository.

Model consumes as input L-channel of LAB-image. Model give as output predict A- and B-channels of LAB-image.

Metric

Value

Type

Colorization

GFLOPs

83.6045

MParams

32.2360

Source framework

PyTorch*

## Accuracy¶

The accuracy metrics were calculated between generated images by model and real validation images from ImageNet dataset. Results are obtained on subset of 2000 images.

Metric

Value

PSNR

26.99dB

SSIM

0.90

Also, metrics can be calculated using VGG16 caffe model and colorization as preprocessing. The results below are obtained on the validation images from ImageNet dataset.

For preprocessing rgb -> gray -> colorization received values:

Metric

Value with preprocessing

Value without preprocessing

Accuracy top-1

57.75%

70.96%

Accuracy top-5

81.50%

89.88%

## Input¶

### Original model¶

Image, name - data_l, shape - 1, 1, 256, 256, format is B, C, H, W, where:

• B - batch size

• C - channel

• H - height

• W - width

Channel order is L-channel.

### Converted model¶

Image, name - data_l, shape - 1, 1, 256, 256, format is B, C, H, W, where:

• B - batch size

• C - channel

• H - height

• W - width

Channel order is L-channel.

## Output¶

### Original model¶

Image, name - color_ab, shape - 1, 2, 256, 256, format is B, C, H, W, where:

• B - batch size

• C - channel

• H - height

• W - width

Channel order is AB channels of LAB-image.

### Converted model¶

Image, name - color_ab, shape - 1, 2, 256, 256, format is B, C, H, W, where:

• B - batch size

• C - channel

• H - height

• W - width

Channel order is AB channels of LAB-image.

You can download models and if necessary convert them into OpenVINO™ IR format using the Model Downloader and other automation tools as shown in the examples below.

omz_downloader --name <model_name>
omz_converter --name <model_name>