netvlad-tf#

Use Case and High-Level Description#

NetVLAD is a CNN architecture which tackles the problem of large scale visual place recognition. The architecture uses VGG 16 as base network and NetVLAD - a new trainable generalized VLAD (Vector of Locally Aggregated Descriptors) layer. It is a place recognition model pre-trained on the Pittsburgh 250k dataset.

For details see repository and paper.

Specification#

Metric

Value

Type

Place recognition

GFLOPs

36.6374

MParams

149.0021

Source framework

TensorFlow*

Accuracy#

Accuracy metrics are obtained on a smaller validation subset of Pittsburgh 250k dataset (Pitts30k) containing 10k database images in each set (train/test/validation). Images were resized to input size.

Metric

Value

localization_recall

82.0321%

Input#

Original model#

Image, name - Placeholder, shape - 1, 200, 300, 3, format is B, H, W, C, where:

  • B - batch size

  • C - channel

  • H - height

  • W - width

Channel order is RGB.

Converted model#

Image, name - Placeholder, shape - 1, 200, 300, 3, format is B, H, W, C, where:

  • B - batch size

  • C - channel

  • H - height

  • W - width

Channel order is BGR.

Output#

Original model#

Floating point embeddings, name - vgg16_netvlad_pca/l2_normalize_1, shape - 1, 4096, output data format - B, C, where:

  • B - batch size

  • C - vector of 4096 floating points values, local image descriptors

Converted model#

Floating point embeddings, name - vgg16_netvlad_pca/l2_normalize_1, shape - 1, 4096, output data format - B, C, where:

  • B - batch size

  • C - vector of 4096 floating points values, local image descriptors

Download a Model and Convert it into OpenVINO™ IR Format#

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.

An example of using the Model Downloader:

omz_downloader --name <model_name>

An example of using the Model Converter:

omz_converter --name <model_name>

Demo usage#

The model can be used in the following demos provided by the Open Model Zoo to show its capabilities: