Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Action Reconstruction Method Based on Sparse Input Signal

A technology of input signals and actions, applied to instruments, other database retrieval, calculation, etc., can solve problems such as high complexity, low efficiency, poor algorithm adaptability and stability, and achieve the effect of uniform similarity

Active Publication Date: 2019-05-24
BEIJING INSTITUTE OF TECHNOLOGYGY
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, this method has high complexity, low efficiency, large average reconstruction error, and poor adaptability and stability of the algorithm.

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
  • Action Reconstruction Method Based on Sparse Input Signal
  • Action Reconstruction Method Based on Sparse Input Signal

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0017] The present invention will be further described below in conjunction with drawings and embodiments.

[0018] A specific embodiment of the motion reconstruction method based on the sparse input signal of the present invention includes the following steps:

[0019] 1) Choose the free and open HDM05 action library as the source of the action database;

[0020] 2) Read all 216 action clip files into the action database, reorganize them in units of frames, and synthesize one frame based on the average value of every 4 consecutive frames, and obtain a total of about 260,000 frames, each with 93 dimensions action database;

[0021] 3) Calculate the corresponding position, rotation angle, rotation angular velocity and rotation angular acceleration of each frame respectively;

[0022] 5) An index structure based on a K-D tree is established for each frame of rotational angular acceleration data;

[0023] 6) In the online state, use the FAB inertial motion capture device to ob...

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 action reconstruction method based on the sparse input signal proposed by the present invention, step 1, extracting multiple types of data related to actions in an offline state, and constructing an original action database; step 2, using motion capture in an online state The technology obtains the motion data of sparse joints as the input signal for motion reconstruction; Step 3, constructs digital similarity to the input signal, and obtains the retrieval result whose numerical difference value satisfies a certain threshold condition; Step 4, analyzes the retrieval result Do time-continuous construction to obtain a continuous search result sequence on the time axis; Step 5, perform geometric similarity construction on the search result sequence obtained in Step 4, and obtain a unified search result sequence with posture logic and numerical similarity; Step 6 . Perform reconstruction cost optimization on the retrieval result sequence obtained in step five, and use the last frame of the retrieval result sequence with the smallest reconstruction cost as the reconstruction result of the current frame.

Description

technical field [0001] The invention relates to computer graphics and statistics, in particular to a low-cost and high-efficiency motion reconstruction method based on sparse input signals, which belongs to the field of computer applications. technical background [0002] An individual soldier is the most basic unit of an army's combat effectiveness. Improving the level of individual soldier training and equipment is conducive to the improvement of the overall combat effectiveness of the army. Motion capture technology has been widely used in gait analysis, biomechanics, ergonomics, human-computer interaction and control, medical analysis and other fields. Motion capture technology is close to the actual needs of individual soldier training. The application of motion capture technology can effectively improve the development of individual soldier equipment. , training effectiveness evaluation and verification, and the scientific and dynamic monitoring capabilities of individ...

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): G06K9/00G06F16/901G06F16/903
CPCG06F16/9027G06F16/903G06V40/20
Inventor 黄天羽崔丽君丁刚毅
Owner BEIJING INSTITUTE OF TECHNOLOGYGY
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products