deeplabv3

Use Case and High-Level Description

DeepLab is a state-of-art deep learning model for semantic image segmentation. For details see paper.

Specification

Metric

Value

Type

Semantic segmentation

GFLOPs

11.469

MParams

23.819

Source framework

TensorFlow*

Accuracy

Metric

Value

mean_iou

66.85%

Input

Original model

Image, name: ImageTensor, shape: 1, 513, 513, 3, format: B, H, W, C, where:

  • B - batch size

  • H - image height

  • W - image width

  • C - number of channels

Expected color order: RGB.

Converted Model

Image, name: mul_1/placeholder_port_1, shape: 1, 3, 513, 513, format: B, C, H, W, where:

  • B - batch size

  • C - number of channels

  • H - image height

  • W - image width

Expected color order: BGR.

Output

Original Model

Integer values in a range [0, 20], which represent an index of a predicted class for each image pixel. Name: ArgMax, shape: 1, 513, 513 in B, H, W format, where:

  • B - batch size

  • H - image height

  • W - image width

Converted Model

Integer values in a range [0, 20], which represent an index of a predicted class for each image pixel. Name: ArgMax/Squeeze, shape: 1, 513, 513 in B, H, W format, where:

  • B - batch size

  • H - image height

  • W - image width

Download a Model and Convert it into Inference Engine Format

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

An example of using the Model Downloader:

python3 <omz_dir>/tools/downloader/downloader.py --name <model_name>

An example of using the Model Converter:

python3 <omz_dir>/tools/downloader/converter.py --name <model_name>

Use Case and High-Level Description

DeepLab is a state-of-art deep learning model for semantic image segmentation. For details see paper.

Specification

Metric

Value

Type

Semantic segmentation

GFLOPs

11.469

MParams

23.819

Source framework

TensorFlow*

Accuracy

Metric

Value

mean_iou

66.85%

Input

Original model

Image, name: ImageTensor, shape: 1, 513, 513, 3, format: B, H, W, C, where:

  • B - batch size

  • H - image height

  • W - image width

  • C - number of channels

Expected color order: RGB.

Converted Model

Image, name: mul_1/placeholder_port_1, shape: 1, 3, 513, 513, format: B, C, H, W, where:

  • B - batch size

  • C - number of channels

  • H - image height

  • W - image width

Expected color order: BGR.

Output

Original Model

Integer values in a range [0, 20], which represent an index of a predicted class for each image pixel. Name: ArgMax, shape: 1, 513, 513 in B, H, W format, where:

  • B - batch size

  • H - image height

  • W - image width

Converted Model

Integer values in a range [0, 20], which represent an index of a predicted class for each image pixel. Name: ArgMax/Squeeze, shape: 1, 513, 513 in B, H, W format, where:

  • B - batch size

  • H - image height

  • W - image width

Download a Model and Convert it into Inference Engine Format

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

An example of using the Model Downloader:

python3 <omz_dir>/tools/downloader/downloader.py --name <model_name>

An example of using the Model Converter:

python3 <omz_dir>/tools/downloader/converter.py --name <model_name>

Legal Information

The original model is distributed under the Apache License, Version 2.0. A copy of the license is provided in <omz_dir>/models/public/licenses/APACHE-2.0-TF-Models.txt.