Toyota Motor Engineering & Manufacturing North America, Inc.

University of Technology Sydney

Dacheng Tao

Danil V Prokhorov

Xue Mei

Chaohui Wang

Zhibin Hong

Zhe Chen

Jun Li

An object tracking device includes a short-term processing portion and a long short-term processing portion that are implemented by circuitry and work in a collaborative manner to track an object. The short-term processing portion includes a filter that tracks the object based on short-term memory and spatiotemporal consistency. The long short-term processing portion performs key-point matching-tracking and estimation based on a key-point database in order to track the object. A controller determines an output of the object tracking device based on the processing conducted by the short-term and long short-term processing portions of the tracking device, respectively.

Apparatus and method for object tracking

A method of tracking a target object in frames of video data includes receiving a first tracking position associated with the target object in a first frame of a video sequence; identifying, for a second frame of the video sequence, a plurality of representation levels and at least one node for each representation level; determining, by a processor, a second tracking position in the second frame by estimating motion of the target object in the second frame between the first frame and the second frame; determining, at each representation level by the processor, a value for each node based on a conditional property of the node in the second frame; and adjusting, by the processor, the second tracking position based on the values determined for each of the nodes and interactions between at least some of the nodes at different representation levels.

Tracking using multilevel representations

Disclosed herein are devices, systems, and methods for detecting the presence and orientation of traffic lane markings. Deep convolutional neural networks are used with convolutional layers and max-pooling layers to generate fully connected nodes. After the convolutional and max-pooling layers, two sublayers are applied, one to determine presence and one to determine geometry. The presence of a lane marking segment as detected by the first sublayer can serve as a gate for the second sublayer by regulating the credit assignment for training the network. Only when the first sublayer predicts actual presence will the geometric layout of the lane marking segment contribute to the training of the overall network. This achieves advantages with respect to accuracy and efficiency and contributes to efficient robust model selection.

Multi-task deep convolutional neural networks for efficient and robust traffic lane detection

Growing community of inventors

Ultimo, Australia

Dacheng Tao