instance-segmentation-security-0050
Use case and High-level description
This model is an instance segmentation network for 80 classes of objects. It is a Mask R-CNN with ResNet50 backbone, FPN and Bottom-Up Augmentation blocks and light-weight RPN.
Example
![instance-segmentation-security-0050.png](https://docs.openvinotoolkit.org/latest/instance-segmentation-security-0050.png)
Specification
METRIC | VALUE |
---|---|
MS COCO val2017 box AP | 31.27% |
MS COCO val2017 mask AP | 27.83% |
Max objects to detect | 100 |
GFlops | 46.602 |
MParams | 30.448 |
Source framework | PyTorch* |
Average Precision (AP) is defined and measured according to standard MS COCO evaluation procedure.
Performance
Inputs
- name:
im_data
, shape: [1x3x480x480] - An input image in the format [1xCxHxW]. The expected channel order is BGR.
- name:
im_info
, shape: [1x3] - Image information: processed image height, processed image width and processed image scale w.r.t. the original image resolution.
Outputs
- name:
classes
, shape: [100, ] - Contiguous integer class ID for every detected object, '0' for background, i.e. no object.
- name:
scores
: shape: [100, ] - Detection confidence scores in range [0, 1] for every object.
- name:
boxes
, shape: [100, 4] - Bounding boxes around every detected objects in (top_left_x, top_left_y, bottom_right_x, bottom_right_y) format.
- name:
raw_masks
, shape: [100, 81, 28, 28] - Segmentation heatmaps for all classes for every output bounding box.
Legal Information
[*] Other names and brands may be claimed as the property of others.