3D Human Models Applied to Pedestrian Pose Classification
AssigneeHonda Motor Co., Ltd.
InventorBernd Heisele
Life of this patent
Estimating the term…
Abstract
A pedestrian pose classification model is trained. A three-dimensional (3D) model of a pedestrian is received. A set of image parameters indicating how to generate an image of a pedestrian is received. A two-dimensional (2D) synthetic image is generated based on the received 3D model and the received set of image parameters. The generated synthetic image is annotated with the set of image parameters. A plurality of pedestrian pose classifiers is trained through the annotated synthetic image.
Claims20
A method for training a pedestrian pose classification model comprising:
- receiving a three-dimensional (3D) model of a pedestrian;
- receiving a set of image parameters indicating how to generate an image of a pedestrian;
- generating a two-dimensional (2D) synthetic image based on the received 3D model and the received set of image parameters;
- annotating the generated synthetic image with the set of image parameters;
- and
- training a plurality of pedestrian pose classifiers through the annotated synthetic image.
A non-transitory computer-readable storage medium configured to store instructions for training a pedestrian pose classification model, the instructions when executed by a processor cause the processor to:
- receive a three-dimensional (3D) model of a pedestrian;
- receive a set of image parameters indicating how to generate an image of a pedestrian;
- generate a two-dimensional (2D) synthetic image based on the received 3D model and the received set of image parameters;
- annotate the generated synthetic image with the set of image parameters;
- and
- train a plurality of pedestrian pose classifiers through the annotated synthetic image.
A system for training a pedestrian pose classification model, comprising:
- a processor;
- and
- a non-transitory computer-readable storage medium storing instructions, the instructions when executed by the processor cause the processor to:
- receive a three-dimensional (3D) model of a pedestrian;
- receive a set of image parameters indicating how to generate an image of a pedestrian;
- generate a two-dimensional (2D) synthetic image based on the received 3D model and the received set of image parameters;
- annotate the generated synthetic image with the set of image parameters;
- and
- train a plurality of pedestrian pose classifiers through the annotated synthetic image.
Dependencies are inferred from the claim text.