Skip to content

Commit 90bf23e

Browse files
LeonidBeynensoneizamalieveaidova
authored
Add vehicle-attributes-recognition-barrier-0042 model (openvinotoolkit#1155)
* Add vehicle-attributes-recognition-barrier-0042 model * Reverted changes in demos/tests/cases.py * Apply suggestions from code review vehicle-attributes-recognition-barrier-0042 Co-authored-by: Eduard Zamaliev <48511626+eizamaliev@users.noreply.github.com> * Minor change in vehicle-attributes-recognition-barrier-0042 docs * Decorative change in vehicle-attributes-recognition-barrier-0042.md Co-authored-by: Eduard Zamaliev <48511626+eizamaliev@users.noreply.github.com> * Update accuracy checker config for vehicle-attributes-recognition-barrier-0042 Co-authored-by: Katya <ekaterina.aidova@intel.com> * Apply suggestions from code review Co-authored-by: Katya <ekaterina.aidova@intel.com> * Fix description of outputs of vehicle-attributes-recognition-barrier-0042 * Update docs for vehicle-attributes-recognition-barrier-0042/description/vehicle-attributes-recognition-barrier-0042 Co-authored-by: Eduard Zamaliev <48511626+eizamaliev@users.noreply.github.com> Co-authored-by: Eduard Zamaliev <48511626+eizamaliev@users.noreply.github.com> Co-authored-by: Katya <ekaterina.aidova@intel.com>
1 parent 15e5db8 commit 90bf23e

File tree

7 files changed

+137
-3
lines changed

7 files changed

+137
-3
lines changed

‎demos/README.md

+1
Original file line numberDiff line numberDiff line change
@@ -77,6 +77,7 @@ The table below shows the correlation between models, demos, and supported plugi
7777
| head-pose-estimation-adas-0001 | [Interactive Face Detection Demo](./interactive_face_detection_demo/README.md) | Supported | Supported | Supported | Supported |
7878
| license-plate-recognition-barrier-0001 | [Security Barrier Camera Demo](./security_barrier_camera_demo/README.md) | Supported | Supported | Supported | Supported |
7979
| vehicle-attributes-recognition-barrier-0039 | [Security Barrier Camera Demo](./security_barrier_camera_demo/README.md) | Supported | Supported | Supported | Supported |
80+
| vehicle-attributes-recognition-barrier-0042 | [Security Barrier Camera Demo](./security_barrier_camera_demo/README.md) | Supported | | | |
8081
| vehicle-license-plate-detection-barrier-0106 | [Security Barrier Camera Demo](./security_barrier_camera_demo/README.md) | Supported | Supported | Supported | Supported |
8182
| vehicle-license-plate-detection-barrier-0123 | [Security Barrier Camera Demo](./security_barrier_camera_demo/README.md) | Supported | Supported | Supported | Supported |
8283
| face-reidentification-retail-0095 | [Smart Classroom Demo](./smart_classroom_demo/README.md)<br>[Interactive Face Recognition Python* Demo](./python_demos/face_recognition_demo/README.md) | Supported | Supported | Supported | Supported |

‎demos/security_barrier_camera_demo/README.md

+1-1
Original file line numberDiff line numberDiff line change
@@ -3,7 +3,7 @@
33
This demo showcases Vehicle and License Plate Detection network followed by the Vehicle Attributes Recognition and License Plate Recognition networks applied on top
44
of the detection results. You can use a set of the following pre-trained models with the demo:
55
* `vehicle-license-plate-detection-barrier-0106` or `vehicle-license-plate-detection-barrier-0123`, which is primary detection network to find the vehicles and license plates
6-
* `vehicle-attributes-recognition-barrier-0039`, which is executed on top of the results from the first network and
6+
* `vehicle-attributes-recognition-barrier-0039` or `vehicle-attributes-recognition-barrier-0042`, which is executed on top of the results from the first network and
77
reports general vehicle attributes, for example, vehicle type (car/van/bus/track) and color
88
* `license-plate-recognition-barrier-0001` or `license-plate-recognition-barrier-0007`, which is executed on top of the results from the first network
99
and reports a string per recognized license plate

‎models/intel/index.md

+1
Original file line numberDiff line numberDiff line change
@@ -70,6 +70,7 @@ Age/Gender recognition after Face Detection).
7070
| [head-pose-estimation-adas-0001](./head-pose-estimation-adas-0001/description/head-pose-estimation-adas-0001.md) | 0.105 | 1.911 |
7171
| [license-plate-recognition-barrier-0001](./license-plate-recognition-barrier-0001/description/license-plate-recognition-barrier-0001.md) | 0.328 | 1.218 |
7272
| [vehicle-attributes-recognition-barrier-0039](./vehicle-attributes-recognition-barrier-0039/description/vehicle-attributes-recognition-barrier-0039.md) | 0.126 | 0.626 |
73+
| [vehicle-attributes-recognition-barrier-0042](./vehicle-attributes-recognition-barrier-0042/description/vehicle-attributes-recognition-barrier-0042.md) | 0.462 | 11.177 |
7374
| [emotions-recognition-retail-0003](./emotions-recognition-retail-0003/description/emotions-recognition-retail-0003.md) | 0.126 | 2.483 |
7475
| [landmarks-regression-retail-0009](./landmarks-regression-retail-0009/description/landmarks-regression-retail-0009.md) | 0.021 | 0.191 |
7576
| [facial-landmarks-35-adas-0002](./facial-landmarks-35-adas-0002/description/facial-landmarks-35-adas-0002.md) | 0.042 | 4.595 |

‎models/intel/vehicle-attributes-recognition-barrier-0039/description/vehicle-attributes-recognition-barrier-0039.md

+2-2
Original file line numberDiff line numberDiff line change
@@ -19,7 +19,7 @@ This model presents a vehicle attributes classification algorithm for a traffic
1919
| Supported types | Car, bus, truck, van |
2020
| GFlops | 0.126 |
2121
| MParams | 0.626 |
22-
| Source framework | Caffe* |
22+
| Source framework | Caffe\* |
2323

2424
## Accuracy - Confusion Matrix
2525

@@ -69,4 +69,4 @@ Expected color order: BGR.
6969
[car, bus, truck, van]
7070

7171
## Legal Information
72-
[*] Other names and brands may be claimed as the property of others.
72+
[\*] Other names and brands may be claimed as the property of others.

‎models/intel/vehicle-attributes-recognition-barrier-0042/description/vehicle-attributes-recognition-barrier-0042-1.png

59.7 KB
Loading

‎models/intel/vehicle-attributes-recognition-barrier-0042/description/vehicle-attributes-recognition-barrier-0042.md

+71
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,71 @@
1+
# vehicle-attributes-recognition-barrier-0042
2+
3+
## Use Case and High-Level Description
4+
5+
This model presents a vehicle attributes classification algorithm for a traffic analysis scenario.
6+
7+
## Example
8+
9+
![](./vehicle-attributes-recognition-barrier-0042-1.png)
10+
11+
## Specification
12+
13+
| Metric | Value |
14+
|-----------------------|----------------------------------------------|
15+
| Car pose | Front facing cars |
16+
| Occlusion coverage | <50% |
17+
| Min object width | 72 pixels |
18+
| Supported colors | White, gray, yellow, red, green, blue, black |
19+
| Supported types | Car, van, truck, bus |
20+
| GFlops | 0.462 |
21+
| MParams | 11.177 |
22+
| Source framework | PyTorch\* |
23+
24+
## Accuracy
25+
26+
### Color accuracy, %
27+
28+
| Color | Accuracy |
29+
|:--------:|:----------:|
30+
| white | 85.61% |
31+
| gray | 77.47% |
32+
| yellow | 49.73% |
33+
| red | 97.62% |
34+
| green | 74.24% |
35+
| blue | 80.02% |
36+
| black | 97.55% |
37+
38+
**Color average accuracy: 80.32%**
39+
40+
### Type accuracy, %
41+
42+
| Type | Accuracy |
43+
|:-----:|:--------:|
44+
| car | 97.96% |
45+
| van | 86.08% |
46+
| truck | 97.47% |
47+
| bus | 42.49% |
48+
49+
**Type average accuracy: 81.00%**
50+
51+
## Performance
52+
53+
## Inputs
54+
55+
Name: `input` , shape: [1x3x72x72] - an input image in following format
56+
[1xCxHxW], where:
57+
- C - number of channels
58+
- H - image height
59+
- W - image width
60+
61+
Expected color order: BGR.
62+
63+
## Outputs
64+
65+
1. Name: `color`, shape: [1, 7] - probabilities across seven color classes
66+
[`white`, `gray`, `yellow`, `red`, `green`, `blue`, `black`]
67+
2. Name: `type`, shape: [1, 4] - probabilities across four type classes
68+
[`car`, `van`, `truck`, `bus`]
69+
70+
## Legal Information
71+
[\*] Other names and brands may be claimed as the property of others.

‎tools/accuracy_checker/configs/vehicle-attributes-recognition-barrier-0042.yml

+61
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,61 @@
1+
models:
2+
- name: vehicle-attributes-recognition-barrier-0042
3+
4+
launchers:
5+
- framework: dlsdk
6+
tags:
7+
- FP32
8+
model: intel/vehicle-attributes-recognition-barrier-0042/FP32/vehicle-attributes-recognition-barrier-0042.xml
9+
weights: intel/vehicle-attributes-recognition-barrier-0042/FP32/vehicle-attributes-recognition-barrier-0042.bin
10+
adapter:
11+
type: vehicle_attributes
12+
color_out: color
13+
type_out: type
14+
15+
- framework: dlsdk
16+
tags:
17+
- FP16
18+
model: intel/vehicle-attributes-recognition-barrier-0042/FP16/vehicle-attributes-recognition-barrier-0042.xml
19+
weights: intel/vehicle-attributes-recognition-barrier-0042/FP16/vehicle-attributes-recognition-barrier-0042.bin
20+
adapter:
21+
type: vehicle_attributes
22+
color_out: color
23+
type_out: type
24+
25+
- framework: dlsdk
26+
tags:
27+
- FP16-INT8
28+
model: intel/vehicle-attributes-recognition-barrier-0042/FP16-INT8/vehicle-attributes-recognition-barrier-0042.xml
29+
weights: intel/vehicle-attributes-recognition-barrier-0042/FP16-INT8/vehicle-attributes-recognition-barrier-0042.bin
30+
adapter:
31+
type: vehicle_attributes
32+
color_out: color
33+
type_out: type
34+
35+
datasets:
36+
- name: vehicle_attributes_0042
37+
38+
preprocessing:
39+
- type: extend_around_rect
40+
augmentation_param: 0.3
41+
- type: crop_rect
42+
- type: resize
43+
size: 115
44+
- type: crop
45+
size: 72
46+
47+
metrics:
48+
- name: color_accuracy
49+
type: accuracy_per_class
50+
presenter: print_vector
51+
annotation_source: color
52+
prediction_source: color
53+
label_map: color_label_map
54+
- name: type_accuracy
55+
type: accuracy_per_class
56+
presenter: print_vector
57+
annotation_source: type
58+
prediction_source: type
59+
label_map: type_label_map
60+
61+
global_definitions: ../dataset_definitions.yml

0 commit comments

Comments
 (0)