A method and device for image model construction
An image model and image technology, applied in image analysis, image enhancement, image data processing, etc., can solve the problems of cumbersome use, high strength, and high cost
Active Publication Date: 2016-12-28
ALCATEL LUCENT SAS
View PDF3 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0003] Dedicated hardware solutions exist but they are expensive, use expensive cameras, and are cumbersome to use
Furthermore, most solutions do not allow scenes to be dynamic, which significantly limits their use
[0004] For three-dimensional, which, in the rest of the text will be abbreviated as 3D, construction from 3D measurements, state-of-the-art meshing algorithms are able to create results with good quality from high-quality measurements; however these solutions are computational very high strength
Furthermore, solutions for generating 3D models with good quality based on lower quality images are not available
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 moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0039] Figure 1a A schematic diagram showing an embodiment of a method for generating and providing a 2D or 3D image model from a single image data input. The image data input may have been provided by a camera providing a still image or a series of pictures representing a moving object, possibly in 2D, 3D or 2D+z format. By 2D+z it is meant that two-dimensional pixel data is provided together with depth data. Such a representation can be used to reconstruct 3D pixel data, and is usually generated by a 3D camera. The image input data may also be obtained eg from a memory or storage device, or provided via any type of communication network, eg MMS pictures sent by conventional mobile phones. Input image data in Figure 1a, denoted as IV1 and subjected to two operations. The first operation involves state extraction or determination by which a state parameter representing the state of the object representing the image input is determined. Configurations of object features ar...
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
Login to View More Abstract
A method for building an image model (M1; M) from at least one image data input (IV1; IV1‑IVn), comprising the steps, in an iterative manner, of determining said at least one image data input (IV1; IV1 -IVn) at least one state (PS1; PS1-PSn), and a state (PSMF) of an intermediate learning model (MF; MIF) -said at least one state (PS1; PS1-PSn) from said at least one image data input ), and determining a target state (TSP) from said state (PSMF) of said intermediate learning model (MF; MIF), - according to said determined target state (TSP), said at least one image data input (IV1; IV1‑IVn) performing at least one transformation resulting in at least one transformed image (IV1T; IV1T‑IVnT), aggregating said at least one transformed image (IV1T; IV1T‑IVnT) with intermediate learning Model (MF; MIF; MIT; MFT) information, thereby generating an updated estimate of said image model (M1; M), - providing said updated estimate of said image model (M1; M) as said image model (M1; M) also - in a feedback loop, providing said updated estimate of said image model (M1; M) to a model object learning module (500) for deriving said intermediate learning model (MF, MIF) update.
Description
technical field [0001] The present invention relates to the field of image model construction. Background technique [0002] Currently, model building based on real dynamic scenes or even based on images captured by cheap cameras can be a challenge. [0003] Dedicated hardware solutions exist but they are expensive, use expensive cameras, and are cumbersome to use. Furthermore, most solutions also do not allow scenes to be dynamic, which significantly limits their use. [0004] For three-dimensional, which, in the rest of the text will be abbreviated as 3D, construction from 3D measurements, state-of-the-art meshing algorithms are able to create results with good quality from high-quality measurements; however these solutions are computational Very high strength. Furthermore, solutions for generating 3D models with good quality based on lower quality images are not available. Contents of the invention [0005] It is therefore an object of embodiments of the present inv...
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
Login to View More Patent Type & Authority Patents(China)
IPC IPC(8): G06T7/20
CPCG06T2207/10016G06T2207/20121G06T2207/30201G06T7/251G06T7/20G06T17/00G06T19/00G06T19/20
Inventor 唐尼·媞加替欧文·塞克斯萨米·利芬斯马滕·阿兹
Owner ALCATEL LUCENT SAS