Vehicle-Human algorithm

Object classification

SentiVeillance allows to perform object classification, locations and tracking based on its type. Besides general classification between pedestrians, cars, busses, bikes and trucks, SentiVeillance can separate advanced vehicle types with over 20 classes available, ranging from common (sedan, SUV, bus, truck etc.) to country-specific or historic (tuk-tuk, carriage etc).

Color estimation

The algorithm returns most likely color estimation for cars and pedestrians. The estimated color values are: red, orange, yellow, green, blue, silver, white, black, brown, grey.

Vehicle make and model estimation

SentiVeillance technology includes a neural networks-based estimation of the tracked vehicle's make (brand) and model. Besides general usage, this data can be combined with the recognized license plate and checked against the data in the vehicles registry.

Clothing analysis

When a pedestrian is detected, their appearance can be evaluated for the presence of different clothes or headwear types. Over 20 clothes and headwear classes are recognized, ranging from common (shirts, trousers, skirts, coats etc.) to national (i.e. sari, niqab, kimono etc.). Also, this algorithm can estimate if the person is partially or totally naked.

Movement direction estimation

The algorithm estimates vehicle or pedestrian movement direction as angle relative to camera orientation.

Tolerance to object visibility

The detection algorithm works with partially visible objects and from great distance.

Object classification

SentiVeillance allows to perform object classification, locations and tracking based on its type. Besides general classification between pedestrians, cars, busses, bikes and trucks, SentiVeillance 8.1 SDK performs advanced vehicle type classification with over 20 classes available, ranging from common (sedan, SUV, bus, truck etc.) to country-specific or historic (tuk-tuk, carriage etc).

Color estimation

The algorithm returns most likely color estimation for cars and pedestrians. The estimated color values are: red, orange, yellow, green, blue, silver, white, black, brown, grey.

Vehicle make and model estimation

SentiVeillance 8.1 SDK includes a neural networks-based estimation of the tracked vehicle’s make (brand) and model. Besides general usage, this data can be combined with the recognized license plate and checked against the data in the vehicles registry.

Clothing analysis

Movement direction estimation

The algorithm estimates vehicle or pedestrian movement direction as angle relative to camera orientation.

Tolerance to object visibility

The detection algorithm works with partially visible objects and from great distance.

Combinations with other algorithms

In addition, data about the detected objects can be used in combination with other SentiVeillance modalities:

Face modality can be used together with the VH modality to perform unknown persons’ identification.
ALPR modality can be used together with the VH modality to perform vehicle identification by its license plate.

Programming samples from the SDK show how to use this modality. A license for using the SentiVeillance VH modality on one video stream is included with SentiVeillance SDK. Additional licenses or upgrade for the number of processed video streams in the existing licenses can be purchased any time by SentiVeillance SDK customers.

SentiVeillance

developed by Neurotechnology

Face

Vehicle-Human

Automatic license plate recognition

Vehicle-Human algorithm

Object classification

Color estimation

Vehicle make and model estimation

Clothing analysis

Movement direction estimation

Tolerance to object visibility

Combinations with other algorithms

Check out other SentiVeillance algorithms

SentiVeillance

developed by Neurotechnology