person-attributes-recognition-crossroad-0234

Use Case and High-Level Description

This model presents a person attributes classification algorithm analysis scenario. The model consists of the ResNet-50 backbone and a head. For an input image with a pedestrian the model returns 7 values that are probabilities of the corresponding 7 attributes.

Specification

Metric Value
Pedestrian pose Standing person
Occlusion coverage <20%
Min object width 80 pixels
Supported attributes is_male, has_bag, has_hat, has_longsleeves, has_longpants, has_longhair, has_coat_jacket
GFlops 2.167
MParams 23.510
Source framework PyTorch*

Accuracy

Attribute F1
is_male 0.92
has_bag 0.44
has_hat 0.74
has_longsleeves 0.45
has_longpants 0.89
has_longhair 0.84
has_coat_jacket NA

Inputs

Image, name: input, shape: 1, 3, 160, 80 in the format 1, C, H, W, where:

  • C - number of channels
  • H - image height
  • W - image width

The expected color order is BGR.

Outputs

The net output is a blob named attributes with shape 1, 7 across seven attributes: [is_male, has_bag, has_hat, has_longsleeves, has_longpants, has_longhair, has_coat_jacket]. Value > 0.5 means that the corresponding attribute is present.

Legal Information

[*] Other names and brands may be claimed as the property of others.