This is an action recognition composite model for the driver monitoring use case, consisting of encoder and decoder parts. The encoder model uses Video Transformer approach with MobileNetv2 encoder. It is able to recognize the following actions: drinking, doing hair or making up, operating the radio, reaching behind, safe driving, talking on the phone, texting.
The driver-action-recognition-adas-0002-encoder model accepts video frame and produces embedding. Video frames should be sampled to cover ~1 second fragment (i.e. skip every second frame in 30 fps video).
name: "0" , shape: [1x3x224x224] - An input image in the format [BxCxHxW], where:
Expected color order is BGR.
The model outputs a tensor with the shape [1x512x1x1], representing embedding of precessed frame.
The driver-action-recognition-adas-0002-decoder model accepts stack of frame embeddings, computed by driver-action-recognition-adas-0002-encoder, and produces prediction on input video.
The model outputs a tensor with the shape [bx9], each row is a logits vector of performed actions.
[*] Other names and brands may be claimed as the property of others.