Processing apparatus, processing system, imaging apparatus, processing method, and storage medium
AssigneeCanon Kabushiki Kaisha
InventorsYuichi Kusumi, Chiaki Inoue, Yoshiaki Ida
Life of this patent
Estimating the term…
Abstract
A processing apparatus includes an input image acquirer configured to acquire an input image obtained by capturing an object, a normal information acquirer configured to acquire normal information of the object, a first renderer configured to generate a first rendered image about the object based on a light source condition in capturing the input image and the normal information, and an area detector configured to detect a target area corresponding to a portion where an error of the normal information in the object is larger than a predetermined value based on the input image and the first rendered image.
Claims20
A processing apparatus comprising:
- an input image acquirer configured to acquire an input image obtained by capturing an object;
- a normal information acquirer configured to acquire normal information of the object;
- a first renderer configured to generate a first rendered image about the object based on a light source condition in capturing the input image and the normal information;
- and
- an area detector configured to detect a target area corresponding to a portion where an error of the normal information in the object is larger than a predetermined value based on the input image and the first rendered image.
A processing system comprising:
- a light source unit;
- and
- a processing apparatus,
- wherein the processing apparatus that includes:
- an input image acquirer configured to acquire an input image obtained by capturing an object;
- a normal information acquirer configured to acquire normal information of the object;
- a first renderer configured to generate a first rendered image about the object based on a light source condition in capturing the input image and the normal information;
- and
- an area detector configured to detect a target area corresponding to a portion where an error of the normal information in the object is larger than a predetermined value based on the input image and the first rendered image.
An imaging apparatus comprising:
- an image sensor configured to photoelectrically convert an optical image formed via an imaging optical system and to output image data corresponding to an input image;
- and
- a processing apparatus,
- wherein the processing apparatus that includes:
- an input image acquirer configured to acquire an input image obtained by capturing an object;
- a normal information acquirer configured to acquire normal information of the object;
- a first renderer configured to generate a first rendered image about the object based on a light source condition in capturing the input image and the normal information;
- and
- an area detector configured to detect a target area corresponding to a portion where an error of the normal information in the object is larger than a predetermined value based on the input image and the first rendered image.
A processing method comprising the steps of:
- acquiring an input image obtained by capturing an object;
- acquiring normal information of the object;
- generating a first rendered image about the object based on a light source condition in capturing the input image and the normal information;
- and
- detecting a target area corresponding to a portion where an error of the normal information in the object is larger than a predetermined value based on the input image and the first rendered image.
A non-transitory computer-readable storage medium storing a program for causing a computer to execute a processing method,
- wherein a processing method includes the steps of:
- acquiring an input image obtained by capturing an object;
- acquiring normal information of the object;
- generating a first rendered image about the object based on a light source condition in capturing the input image and the normal information;
- and
- detecting a target area corresponding to a portion where an error of the normal information in the object is larger than a predetermined value based on the input image and the first rendered image.
Dependencies are inferred from the claim text.