Check patentability & draft patents in minutes with Patsnap Eureka AI!

Light Field Depth Estimation Method Based on Orientation and Scale Adaptive Convolutional Neural Network

A convolutional neural network, scale adaptive technology, applied in biological neural network model, neural architecture, computing and other directions, can solve the problem of difficult to calculate the depth of the occluded area in the textureless area of ​​the image

Active Publication Date: 2022-03-01
HANGZHOU DIANZI UNIV
View PDF14 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, this method also faces problems such as difficulty in calculating the depth of texture-free areas and occluded areas of the image.

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Light Field Depth Estimation Method Based on Orientation and Scale Adaptive Convolutional Neural Network
  • Light Field Depth Estimation Method Based on Orientation and Scale Adaptive Convolutional Neural Network
  • Light Field Depth Estimation Method Based on Orientation and Scale Adaptive Convolutional Neural Network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0054] Below in conjunction with accompanying drawing and specific embodiment the present invention is described in further detail:

[0055] Such as Figure 1-4 As shown, a new light field depth estimation method based on direction and scale adaptive convolutional neural network is used for multi-directional and scale light field data, including the following process:

[0056] The concrete of the inventive method comprises following process:

[0057] The technical scheme that the present invention solves its technical problem to take comprises the steps:

[0058] Step 1. Prepare the light field data set, make training set and test set;

[0059] Step 2. Build a direction- and scale-adaptive convolutional neural network SOA-EPN;

[0060] Step 3. Use the training set to train the built SOA-EPN network;

[0061] Step 4. Use the trained SOA-EPN network to test on the test set;

[0062] Step 1 specifically includes the following steps:

[0063] Step 1-1: Use the 4D light field...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a method for estimating the depth of a light field based on a direction- and scale-adaptive convolutional neural network. The present invention comprises the following steps: Step 1. Prepare a light field data set, make a training set and a test set; Step 2. Build a direction- and scale-adaptive convolutional neural network SOA‑EPN; Step 3. Use the training set to train and build SOA-EPN network; Step 4. Use the practiced SOA-EPN network to test on the test set; the present invention predicts the depth of light field by means of scale and direction-aware convolutional neural networks, which not only utilizes multiple directions, but also has a good The problems such as occlusion are dealt with, and accurate depth estimation results are obtained.

Description

technical field [0001] The invention relates to the field of deep learning and light field depth estimation, in particular to a light field depth estimation method based on direction and scale adaptive convolutional neural networks. Background technique [0002] Deep learning is an important breakthrough in the field of artificial intelligence in recent years, and has made breakthroughs in image recognition, speech recognition, natural language processing and other directions. Compared with traditional machine learning methods, the main process of deep learning methods is: constructing data sets; using convolutional layers, fully connected layers, activation layers, etc. to define deep neural networks and define loss functions; using data sets to train defined deep networks Model, which uses optimization methods such as backpropagation techniques and gradient descent to update the parameters of the neural network. The trained deep network model can fit a high-dimensional co...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Patents(China)
IPC IPC(8): G06T7/557G06N3/04
CPCG06T7/557G06N3/045G06N3/044
Inventor 周文晖梁麟开魏兴明周恩慈
Owner HANGZHOU DIANZI UNIV
Features
  • R&D
  • Intellectual Property
  • Life Sciences
  • Materials
  • Tech Scout
Why Patsnap Eureka
  • Unparalleled Data Quality
  • Higher Quality Content
  • 60% Fewer Hallucinations
Social media
Patsnap Eureka Blog
Learn More

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2025 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com