Monocular cued detection of three-dimensional strucures from depth images

A navigation system for a host vehicle, the system comprising:

• at least one processor configured to:
• receive a plurality of image frames captured by a camera from an environment of the host vehicle;
• detect, from the plurality of image frames, an imaged feature from the environment of the host vehicle;
• generate, based on the plurality of image frames, a depth map of the imaged feature, the depth map representing a distance from the camera to the image feature;
• compare the depth map with one or more reference maps to determine whether the imaged feature is an obstacle to a motion of the host vehicle;
• and
• cause implementation of a navigational action based on a result of the determination that the imaged feature is an obstacle.

A method for navigating a host vehicle, the method comprising:

• receiving a plurality of image frames captured by a camera from an environment of the host vehicle;
• detecting, from the plurality of image frames, an imaged feature from the environment of the host vehicle;
• generating, based on the plurality of image frames, a depth map of the imaged feature, the depth map representing a distance from the camera to the image feature;
• comparing the depth map with one or more reference maps to determine whether the imaged feature is an obstacle to a motion of the host vehicle;
• and
• causing implementation of a navigational action based on a result of the determination that the imaged feature is an obstacle.

A non-transitory computer-readable medium storing instructions that, when executed by a processing device, cause the processing device to perform a method comprising:

• receiving a plurality of image frames captured by a camera from an environment of a host vehicle;
• detecting, from the plurality of image frames, an imaged feature from the environment of the host vehicle;
• generating, based on the plurality of image frames, a depth map of the imaged feature, the depth map representing a distance from the camera to the image feature;
• comparing the depth map with one or more reference maps to determine whether the imaged feature is an obstacle to a motion of the host vehicle;
• and
• causing implementation of a navigational action based on a result of the determination that the imaged feature is an obstacle.