Overview of OpenVINO™ Toolkit Intel's Pre-Trained Models

OpenVINO™ toolkit provides a set of pre-trained models that you can use for learning and demo purposes or for developing deep learning software. Most recent version is available in the repo on Github.

The models can be downloaded via Model Downloader (<OPENVINO_INSTALL_DIR>/deployment_tools/open_model_zoo/tools/downloader). They can also be downloaded manually from 01.org.

TIP: You also can download and profile Intel® pretrained models inside the OpenVINO™ Deep Learning Workbench (DL Workbench). DL Workbench is a platform built upon OpenVINO™ and provides a web-based graphical environment that enables you to optimize, fine-tune, analyze, visualize, and compare performance of deep learning models on various Intel® architecture configurations. In the DL Workbench, you can use most of OpenVINO™ toolkit components.
Proceed to an easy installation from Docker to get started.

## Object Detection Models

Several detection models can be used to detect a set of the most popular objects - for example, faces, people, vehicles. Most of the networks are SSD-based and provide reasonable accuracy/performance trade-offs. Networks that detect the same types of objects (for example, face-detection-adas-0001 and face-detection-retail-0004) provide a choice for higher accuracy/wider applicability at the cost of slower performance, so you can expect a "bigger" network to detect objects of the same type better.

Model Name Complexity (GFLOPs) Size (Mp) Face Person Vehicle Bike License plate Product
faster-rcnn-resnet101-coco-sparse-60-0001 364.21 52.79 X X X
face-detection-retail-0004 1.067 0.588 X
face-detection-retail-0005 0.982 1.021 X
face-detection-0200 0.785 1.828 X
face-detection-0202 1.767 1.842 X
face-detection-0204 2.405 1.851 X
face-detection-0205 2.853 2.392 X
face-detection-0206 339.597 69.920 X
person-detection-retail-0002 12.427 3.244 X
person-detection-retail-0013 2.300 0.723 X
person-detection-action-recognition-0005 7.140 1.951 X
person-detection-action-recognition-0006 8.225 2.001 X
person-detection-action-recognition-teacher-0002 7.140 1.951 X
person-detection-raisinghand-recognition-0001 7.138 1.951 X
person-detection-0200 0.786 1.817 X
person-detection-0201 1.768 1.817 X
person-detection-0202 3.143 1.817 X
person-detection-0106 404.264 71.565 X
vehicle-detection-0200 0.786 1.817 X
vehicle-detection-0201 1.768 1.817 X
vehicle-detection-0202 3.143 1.817 X
person-vehicle-bike-detection-crossroad-0078 3.964 1.178 X X X
person-vehicle-bike-detection-crossroad-1016 3.560 2.887 X X X
person-vehicle-bike-detection-crossroad-yolov3-1020 65.984 61.922 X X X
person-vehicle-bike-detection-2000 0.787 1.821 X X X
person-vehicle-bike-detection-2001 1.770 1.821 X X X
person-vehicle-bike-detection-2002 3.163 1.821 X X X
product-detection-0001 3.598 3.212 X
person-detection-asl-0001 0.986 1.338 X
yolo-v2-ava-0001 29.38 48.29 X X X
yolo-v2-ava-sparse-35-0001 29.38 48.29 X X X
yolo-v2-ava-sparse-70-0001 29.38 48.29 X X X
yolo-v2-tiny-ava-0001 6.975 15.12 X X X
yolo-v2-tiny-ava-sparse-30-0001 6.975 15.12 X X X
yolo-v2-tiny-ava-sparse-60-0001 6.975 15.12 X X X
yolo-v2-tiny-vehicle-detection-0001 5.424 11.229 X

## Object Recognition Models

Object recognition models are used for classification, regression, and character recognition. Use these networks after a respective detector (for example, Age/Gender recognition after Face Detection).

Model Name Complexity (GFLOPs) Size (Mp)
age-gender-recognition-retail-0013 0.094 2.138
vehicle-attributes-recognition-barrier-0039 0.126 0.626
vehicle-attributes-recognition-barrier-0042 0.462 11.177
emotions-recognition-retail-0003 0.126 2.483
landmarks-regression-retail-0009 0.021 0.191

## Reidentification Models

Precise tracking of objects in a video is a common application of Computer Vision (for example, for people counting). It is often complicated by a set of events that can be described as a "relatively long absence of an object". For example, it can be caused by occlusion or out-of-frame movement. In such cases, it is better to recognize the object as "seen before" regardless of its current position in an image or the amount of time passed since last known position.

The following networks can be used in such scenarios. They take an image of a person and evaluate an embedding - a vector in high-dimensional space that represents an appearance of this person. This vector can be used for further evaluation: images that correspond to the same person will have embedding vectors that are "close" by L2 metric (Euclidean distance).

There are multiple models that provide various trade-offs between performance and accuracy (expect a bigger model to perform better).

Model Name Complexity (GFLOPs) Size (Mp) Rank-1 on Market-1501
person-reidentification-retail-0288 0.174 0.183 86.1%
person-reidentification-retail-0287 0.564 0.595 92.9%
person-reidentification-retail-0286 1.170 1.234 94.8%
person-reidentification-retail-0277 1.993 2.103 96.2%

## Semantic Segmentation Models

Semantic segmentation is an extension of object detection problem. Instead of returning bounding boxes, semantic segmentation models return a "painted" version of the input image, where the "color" of each pixel represents a certain class. These networks are much bigger than respective object detection networks, but they provide a better (pixel-level) localization of objects and they can detect areas with complex shape (for example, free space on the road).

Model Name Complexity (GFLOPs) Size (Mp)
unet-camvid-onnx-0001 260.1 31.03
icnet-camvid-ava-0001 151.82 25.45
icnet-camvid-ava-sparse-30-0001 151.82 25.45
icnet-camvid-ava-sparse-60-0001 151.82 25.45

## Instance Segmentation Models

Instance segmentation is an extension of object detection and semantic segmentation problems. Instead of predicting a bounding box around each object instance instance segmentation model outputs pixel-wise masks for all instances.

Model Name Complexity (GFLOPs) Size (Mp)
instance-segmentation-security-1025 30.146 26.69
instance-segmentation-security-0050 46.602 30.448
instance-segmentation-security-0083 365.626 143.444
instance-segmentation-security-0010 899.568 174.568

## Human Pose Estimation Models

Human pose estimation task is to predict a pose: body skeleton, which consists of keypoints and connections between them, for every person in an input image or video. Keypoints are body joints, i.e. ears, eyes, nose, shoulders, knees, etc. There are two major groups of such metods: top-down and bottom-up. The first detects persons in a given frame, crops or rescales detections, then runs pose estimation network for every detection. These methods are very accurate. The second finds all keypoints in a given frame, then groups them by person instances, thus faster than previous, because network runs once.

Model Name Complexity (GFLOPs) Size (Mp)
human-pose-estimation-0001 15.435 4.099

## Image Processing

Deep Learning models find their application in various image processing tasks to increase the quality of the output.

Model Name Complexity (GFLOPs) Size (Mp)
single-image-super-resolution-1032 11.654 0.030
single-image-super-resolution-1033 16.062 0.030
text-image-super-resolution-0001 1.379 0.003

## Text Detection

Deep Learning models for text detection in various applications.

Model Name Complexity (GFLOPs) Size (Mp)
text-detection-0003 51.256 6.747
text-detection-0004 23.305 4.328
horizontal-text-detection-0001 7.718 2.259

## Text Recognition

Deep Learning models for text recognition in various applications.

Model Name Complexity (GFLOPs) Size (Mp)
text-recognition-0012 1.485 5.568
handwritten-score-recognition-0003 0.792 5.555
handwritten-japanese-recognition-0001 117.136 15.31
formula-recognition-medium-scan-0001:
encoder
decoder

16.56
1.86

1.69
2.56

## Text Spotting

Deep Learning models for text spotting (simultaneous detection and recognition).

Model Name Complexity (GFLOPs) Size (Mp)
text-spotting-0002
text-spotting-0002-detector 185.169 26.497
text-spotting-0002-recognizer-encoder 2.082 1.328
text-spotting-0002-recognizer-decoder 0.002 0.273

## Action Recognition Models

Action Recognition models predict action that is being performed on a short video clip (tensor formed by stacking sampled frames from input video). Some models (for example driver-action-recognition-adas-0002 may use precomputed high-level spatial or spatio-temporal) features (embeddings) from individual clip fragments and then aggregate them in a temporal model to predict a vector with classification scores. Models that compute embeddings are called encoder, while models that predict an actual labels are called decoder.

Model Name Complexity (GFLOPs) Size (Mp)
action-recognition-0001
action-recognition-0001-encoder 7.340 21.276
action-recognition-0001-decoder 0.147 4.405
asl-recognition-0004 6.660 4.133
weld-porosity-detection-0001 3.636 11.173

## Image Retrieval

Deep Learning models for image retrieval (ranking 'gallery' images according to their similarity to some 'probe' image).

Model Name Complexity (GFLOPs) Size (Mp)
image-retrieval-0001 0.613 2.535

## Compressed models

Deep Learning compressed models

Model Name Complexity (GFLOPs) Size (Mp)
resnet50-binary-0001 1.002 7.446
resnet18-xnor-binary-onnx-0001 - -

Model Name Complexity (GFLOPs) Size (Mp)