# yolo-v3-tiny-tf¶

## Use Case and High-Level Description¶

YOLO v3 Tiny is a real-time object detection model implemented with Keras* from this repository and converted to TensorFlow* framework. This model was pre-trained on Common Objects in Context (COCO) dataset with 80 classes.

## Conversion¶

1. Download or clone the original repository (tested on d38c3d8 commit).

2. Use the following commands to get original model (named yolov3_tiny in repository) and convert it to Keras* format (see details in the README.md file in the official repository):

wget -O weights/yolov3-tiny.weights https://pjreddie.com/media/files/yolov3-tiny.weights
2. Convert model weights to Keras*:  python tools/model_converter/convert.py cfg/yolov3-tiny.cfg weights/yolov3-tiny.weights weights/yolov3-tiny.h5 

3. Convert model to protobuf:

python tools/model_converter/keras_to_tensorflow.py --input_model weights/yolov3-tiny.h5 --output_model=weights/yolo-v3-tiny.pb

Metric

Value

Type

Detection

GFLOPs

5.582

MParams

8.848

Source framework

Keras*

## Accuracy¶

Accuracy metrics obtained on Common Objects in Context (COCO) validation dataset for converted model.

Metric

Value

mAP

35.9%

COCO mAP

39.7%

## Input¶

### Original model¶

Image, name - image_input, shape - 1, 416, 416, 3, format is B, H, W, C, where:

• B - batch size

• H - height

• W - width

• C - channel

Channel order is RGB. Scale value - 255.

### Converted model¶

Image, name - image_input, shape - 1, 416, 416, 3, format is B, H, W, C, where:

• B - batch size

• H - height

• W - width

• C - channel

Channel order is BGR.

## Output¶

### Original model¶

1. The array of detection summary info, name - conv2d_9/BiasAdd, shape - 1, 13, 13, 255. The anchor values are 81,82, 135,169, 344,319.

2. The array of detection summary info, name - conv2d_12/BiasAdd, shape - 1, 26, 26, 255. The anchor values are 23,27, 37,58, 81,82.

For each case format is B, Cx, Cy, N\*85, where:

• B - batch size

• Cx, Cy - cell index

• N - number of detection boxes for cell

Detection box has format [x, y, h, w, box_score, class_no_1, …, class_no_80], where:

• (x, y) - raw coordinates of box center, apply sigmoid function to get relative to the cell coordinates

• h, w - raw height and width of box, apply exponential function and multiply by corresponding anchors to get absolute height and width values

• box_score - confidence of detection box, apply sigmoid function to get confidence in [0, 1] range

• class_no_1, …, class_no_80 - probability distribution over the classes in logits format, apply sigmoid function and multiply by obtained confidence value to get confidence of each class

The model was trained on Common Objects in Context (COCO) dataset version with 80 categories of object. Mapping to class names provided in <omz_dir>/data/dataset_classes/coco_80cl.txt file.

### Converted model¶

1. The array of detection summary info, name - conv2d_9/BiasAdd/YoloRegion, shape - 1, 13, 13, 255. The anchor values are 81,82, 135,169, 344,319.

2. The array of detection summary info, name - conv2d_12/BiasAdd/YoloRegion, shape - 1, 26, 26, 255. The anchor values are 23,27, 37,58, 81,82.

For each case format is B, Cx, Cy, N\*85, where:

• B - batch size

• N - number of detection boxes for cell

• Cx, Cy - cell index

Detection box has format [x, y, h, w, box_score, class_no_1, …, class_no_80], where:

• (x, y) - coordinates of box center relative to the cell

• h, w - raw height and width of box, apply exponential function and multiply by corresponding anchors to get absolute height and width values

• box_score - confidence of detection box in [0, 1] range

• class_no_1, …, class_no_80 - probability distribution over the classes in the [0, 1] range, multiply by confidence value to get confidence of each class

The model was trained on Common Objects in Context (COCO) dataset version with 80 categories of object. Mapping to class names provided in <omz_dir>/data/dataset_classes/coco_80cl.txt file.

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>