Joseph Allen Roth

US Patent:

20200117887, Apr 16, 2020

Filed:

Oct 16, 2018

Appl. No.:

16/161927

Inventors:

- Mountain View CA, US
Ondrej Klejch - Mountain View CA, US
Joseph Edward Roth - Pittsburgh PA, US

International Classification:

G06K 9/00
G10L 15/24
G10L 25/24
G06N 3/04
G06N 3/08
G06N 5/04

Abstract:

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for generating predictions for whether a target person is speaking during a portion of a video. In one aspect, a method includes obtaining one or more images which each depict a mouth of a given person at a respective time point. The images are processed using an image embedding neural network to generate a latent representation of the images. Audio data corresponding to the images is processed using an audio embedding neural network to generate a latent representation of the audio data. The latent representation of the images and the latent representation of the audio data is processed using a recurrent neural network to generate a prediction for whether the given person is speaking.

US Patent:

20180272941, Sep 27, 2018

Filed:

Mar 27, 2017

Appl. No.:

15/470159

Inventors:

- Detroit MI, US
- East Lansing MI, US
Xiaoming Liu - Okemos MI, US
Yousef Atoum - Lansing MI, US
Joseph Roth - Pittsburgh PA, US

International Classification:

B60R 1/00
B62D 15/02
G06T 7/73
G06K 9/46
G06K 9/00
G06N 3/08

Abstract:

A method and apparatus for providing trailer information are provided. The method includes detecting a p coupler of the trailer in the image of the rear-facing camera; detecting a positon of the coupler in the received image; and determining a distance between the detected position of the coupler of the trailer and a hitch of vehicle. The method may be use to display information about a trailer coupler or guide a vehicle to line up a vehicle hitch to a trailer coupler.