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: [1x513x513x3], format: [BxHxWxC], 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: [1x3x513x513], format: [BxCxHxW], 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: [1x513x513] in [BxHxW] 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: [1x513x513] in [BxHxW] 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 APACHE-2.0-TF-Models.txt.