Skip to content

Latest commit

 

History

History
94 lines (66 loc) · 3.85 KB

rfcn-resnet101-coco-tf.md

File metadata and controls

94 lines (66 loc) · 3.85 KB
Raw

rfcn-resnet101-coco-tf

Use Case and High-Level Description

R-FCN Resnet-101 model, pretrained on COCO* dataset. Used for object detection. For details, see the paper.

Specification

Metric Value
Type Object detection
GFlops 53.462
MParams 171.85
Source framework TensorFlow*

Accuracy

Metric Value
coco_precision 28.40%
mAP 45.02%

Input

Original Model

Image, name: image_tensor, shape: [1x600x600x3], format: [BxHxWxC], where:

- B - batch size
- H - image height
- W - image width
- C - number of channels

Expected color order: RGB.

Converted Model

  1. Image, name: image_tensor, shape: [1x3x600x600], format: [BxCxHxW], where:

    • B - batch size
    • C - number of channels
    • H - image height
    • W - image width

Expected color order: BGR.

  1. Information of input image size, name: image_info, shape: [1x3], format: [BxC], where:

    • B - batch size
    • C - vector of 3 values in format [H,W,S], where H is an image height, W is an image width, S is an image scale factor (usually 1)

Output

Original Model

  1. Classifier, name: detection_classes. Contains predicted bounding boxes classes in a range [1, 91]. The model was trained on the Microsoft* COCO dataset version with 91 categories of objects, 0 class is for background. Mapping to class names provided in <omz_dir>/data/dataset_classes/coco_91cl_bkgr.txt file.
  2. Probability, name: detection_scores. Contains probability of detected bounding boxes.
  3. Detection box, name: detection_boxes. Contains detection boxes coordinates in a format [y_min, x_min, y_max, x_max], where (x_min, y_min) are coordinates of the top left corner, (x_max, y_max) are coordinates of the right bottom corner. Coordinates are rescaled to input image size.
  4. Detections number, name: num_detections. Contains the number of predicted detection boxes.

Converted Model

The array of summary detection information, name: detection_output, shape: [1, 1, N, 7], where N is the number of detected bounding boxes. For each detection, the description has the format: [image_id, label, conf, x_min, y_min, x_max, y_max], where: - image_id - ID of the image in the batch - label - predicted class ID in range [1, 91], mapping to class names provided in <omz_dir>/data/dataset_classes/coco_91cl_bkgr.txt file - conf - confidence for the predicted class - (x_min, y_min) - coordinates of the top left bounding box corner (coordinates stored in normalized format, in range [0, 1]) - (x_max, y_max) - coordinates of the bottom right bounding box corner (coordinates stored in normalized format, in range [0, 1])

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.