Methods and systems for converting a line drawing to a rendered image

Patent metadata

Field	Value
Publication number	US-10521700-B2
Application number	US-201715842225-A
Country	US
Kind code	B2
Filing date	Dec 14, 2017
Priority date	Dec 14, 2017
Publication date	Dec 31, 2019
Grant date	Dec 31, 2019

Abstract

The system includes a memory that stores instructions for executing processes converting line drawings to rendered images. The system also includes a processor configured to execute the instructions. The instructions cause the processor to: train a neural network to account for irregularities in the line drawings by introducing noise data into training data of the neural network; receive a first line drawing from an input device; generate a first rendered image based on features identified in the first line drawing; and display the first rendered image on an output device.

First claim

What is claimed is:

1. A system comprising:
a memory that stores instructions for executing processes for converting line drawings to rendered images; and
a processor configured to execute the instructions, wherein the instructions cause the processor to:
train a neural network to account for irregularities in the line drawings by introducing noise data into training data of the neural network;
receive a first line drawing from an input device;
generate a first rendered image based on features identified in the first line drawing; and
display the first rendered image on an output device.

2. The system of claim 1 , wherein the neural network is a conditional Generative Adversarial Network (cGAN).

3. The system of claim 1 , wherein the noise data comprises different types of edge noise.

4. The system of claim 3 , wherein the different types of edge noise comprises high-frequency noise, low-frequency noise, and repetitive noise.

5. The system of claim 1 , wherein the processor is further configured to train the neural network using a database of images, and wherein the first line drawing is a hand-drawn image.

6. The system of claim 5 , wherein generating the first rendered image comprises compiling one or more digital images that match respective features identified in the first line drawing.

7. The system of claim 5 , wherein the hand-drawn image is a partial image, and generating the rendered image comprises inserting omitted elements of the partial image based on relationships learned from the database of images.

8. A method comprising:
training a neural network to account for irregularities in line drawings by introducing noise data into training data of the neural network;
receiving a first line drawing from an input device;
generating a first rendered image based on features identified in the first line drawing; and
displaying the first rendered image on an output device.

9. The method of claim 8 , wherein the neural network is a conditional Generative Adversarial Network (cGAN).

10. The method of claim 8 , wherein the noise data comprises different types of edge noise.

11. The method of claim 10 , wherein the different types of edge noise comprises high-frequency noise, low-frequency noise, and repetitive noise.

12. The method of claim 8 , further comprising training the neural network using a database of images, and wherein the first line drawing is a hand-drawn image.

13. The method of claim 12 , wherein generating the first rendered image comprises compiling one or more digital images that match respective features identified in the first line drawing.

14. The method of claim 12 , wherein the hand-drawn image is a partial image, and generating the rendered image comprises inserting omitted elements of the partial image based on relationships learned from the database of images.

15. A non-transitory computer-readable storage medium containing executable computer program code, the code comprising instructions configured to cause a processor to:
train a neural network to account for irregularities in line drawings by introducing noise data into training data of the neural network;
receive a first line drawing from an input device;
generate a first rendered image based on features identified in the first line drawing; and
display the first rendered image on an output device.

16. The medium of claim 15 , wherein the neural network is a conditional Generative Adversarial Network (cGAN).

17. The medium of claim 15 , wherein the noise data comprises different types of edge noise including high-frequency noise, low-frequency noise, and repetitive noise.

18. The medium of claim 15 , wherein the processor is further configured to train the neural network using a database of images, and wherein the first line drawing is a hand-drawn image.

19. The medium of claim 18 , wherein generating the first rendered image comprises compiling one or more digital images that match respective features identified in the first line drawing.

20. The medium of claim 18 , wherein the hand-drawn image is a partial image, and generating the rendered image comprises inserting omitted elements of the partial image based on relationships learned from the database of images.

Assignees

• Honda Motor Co Ltd

Inventors

• Murrish Robert Wesley

Classifications

• G06V10/82 Primary

  using neural networks · CPC title

• G06F18/217

  Validation; Performance evaluation; Active pattern learning techniques · CPC title

• G06N3/045

  Combinations of networks · CPC title

• G06N3/047

  Probabilistic or stochastic networks · CPC title

• G06N3/08

  Learning methods · CPC title