CN109215054A - Face tracking method and system - Google Patents
Face tracking method and system Download PDFInfo
- Publication number
- CN109215054A CN109215054A CN201710516233.0A CN201710516233A CN109215054A CN 109215054 A CN109215054 A CN 109215054A CN 201710516233 A CN201710516233 A CN 201710516233A CN 109215054 A CN109215054 A CN 109215054A
- Authority
- CN
- China
- Prior art keywords
- face
- tracking
- optical flow
- frame
- face frame
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/20—Analysis of motion
- G06T7/246—Analysis of motion using feature-based methods, e.g. the tracking of corners or segments
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/20—Analysis of motion
- G06T7/269—Analysis of motion using gradient-based methods
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Multimedia (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Image Analysis (AREA)
Abstract
The present invention relates to computer vision tracking technique fields, specifically disclose a kind of face tracking method, comprising the following steps: step 1, uniformly spread in the face frame of initial position a little, initialization needs the point set tracked, forms initial frame;Step 2, optical flow tracking is carried out to all the points in face frame using optical flow method, obtains each new position for being tracked point subsequent time;Step 3, the position of subsequent time face frame is gone out based on the new position estimation, while can estimate the size of subsequent time face frame further according to the position of future position, obtains the variation of face scale.The present invention have so that face tracking process more robust with it is practical, improve the speed and accuracy of face tracking, while low to resource allocation request, the beneficial effect reduced costs.
Description
Technical field
The present invention relates to technical field of computer vision, in particular to a kind of face tracking method and system.
Background technique
Face tracking is as an important technical in individual identification, due to its intuitive, at present increasingly by
To attention, while its application is also more and more extensive.
Face tracking process is mainly concerned with the technology of two aspects: human face analysis and target following.Firstly the need of scheming
The region that face is oriented as in, is then based on the individual features of the extracted region face, and carry out the process of target following.Mesh
Preceding existing face tracking technology is broadly divided into two classifications: tracking based on face key point and based on human face region image
Tracking.The first tracking is based primarily upon the key point extracted on face, carries out characteristic point to these key points
Tracking, to realize the tracking of face.This method needs after orienting face, further extracts the key point of face,
Then it is tracked again, more links in step, the disadvantage is that speed is slow.Another kind based on human face region image with
Track method, essence are exactly then to be tracked face as a kind of unified object category.Usual such methods are based on
Line learns the process for completing tracking by constantly updating Optimized model online to realize, the method with respect to before is more multiple
It is miscellaneous, it is therefore desirable to more computing resources, the disadvantage is that more demanding to computing resource.
Based on this, the present invention proposes a kind of fast human face tracking method that may operate on embedded low configuration board, with
Long-term tracking work is carried out to Given Face suitable for the numerous areas such as robot.
Summary of the invention
The present invention is directed to overcome existing face tracking system, calculating speed is slow, the demanding technological deficiency of computing resource, mentions
For a kind of face tracking method and system.
To achieve the above object, the invention adopts the following technical scheme:
The present invention provides a kind of face tracking method, comprising the following steps:
Step 1, it is uniformly spread in the face frame of initial position a little, initialization needs the point set tracked, forms initial frame;
Step 2, optical flow tracking is carried out to all the points in face frame using optical flow method, it is lower for the moment obtains each tracked point
The new position carved;
Step 3, the position of subsequent time face frame is gone out based on the new position estimation, while further according to the position of future position
It can estimate the size of subsequent time face frame, obtain the variation of face scale.
In some embodiments, in the step 2, the optical flow tracking includes preceding to light stream and backward light stream, for rejecting
Track the light stream point of mistake.
In some embodiments, the step 2 includes: the point set of the initial position by preceding new to acquisition after optical flow tracking
The point set of position;To optical flow tracking after being carried out to the point set of new position;Statistics obtains the error of point set tracking.
It is further comprising the steps of in some embodiments: to carry out closest comparison, intercept the face frame region of former frame and work as
The face frame region of previous frame carries out the similarity-rough set of feature, judges whether tracked face frame region is correct.
In some embodiments, closest comparison is carried out using pixel average, variance and EOH feature.
Correspondingly, it the present invention also provides a kind of face tracking system, comprises the following modules:
Face frame setup module, for uniformly spreading in the face frame of initial position a little, initialization needs the point set tracked,
Form initial frame;
Optical flow tracking module obtains each quilt for carrying out optical flow tracking to all the points in face frame using optical flow method
The new position of trace point subsequent time;
Face frame estimation block goes out the position of subsequent time face frame based on the new position estimation, while further according to pre-
The position of measuring point can estimate the size of subsequent time face frame, obtain the variation of face scale.
In some embodiments, to optical flow tracking and backward optical flow tracking before the optical flow tracking module progress, for rejecting
Track the light stream point of mistake.
In some embodiments, the optical flow tracking module, to light stream before being also used to carry out the point set of the initial position
Tracking, obtains the point set of new position;To optical flow tracking after being carried out to the point set of new position;Statistics obtains the error of point set tracking.
In some embodiments, face tracking system further includes closest comparison module, for intercepting the face frame of former frame
The face frame region of region and present frame carries out the similarity-rough set of feature, judges whether tracked face frame region is correct.
In some embodiments, the closest comparison module is carried out most adjacent using pixel average, variance and EOH feature
It is close to compare.
The beneficial effects of the present invention are: face tracking method of the invention by the basis of existing optical flow algorithm into
Row improves, by increasing reverse light stream and closest comparison so that face tracking process more robust with it is practical, improve people
The speed and accuracy of face tracking, at the same it is low to resource allocation request, it reduces costs.
Detailed description of the invention
Fig. 1 is the flow chart of the present inventor's face tracing method;
Fig. 2 is that the face tracking face frame of the present inventor's face tracing method spreads a schematic diagram;
Fig. 3 is forward and reverse light stream schematic diagram in the present inventor's face tracing method;
Fig. 4 is the schematic diagram of EOH feature coding in the present inventor's face tracing method.
Specific embodiment
In order to enable those skilled in the art to better understand the solution of the present invention, below in conjunction in the embodiment of the present invention
Attached drawing, technical scheme in the embodiment of the invention is clearly and completely described, it is clear that described embodiment is only
The embodiment of a part of the invention, instead of all the embodiments.Based on the embodiments of the present invention, ordinary skill people
The model that the present invention protects all should belong in member's every other embodiment obtained without making creative work
It encloses.
Description and claims of this specification and term " first ", " second ", " third " " in above-mentioned attached drawing
The (if present)s such as four " are to be used to distinguish similar objects, without being used to describe a particular order or precedence order.It should manage
The data that solution uses in this way are interchangeable under appropriate circumstances, so that the embodiments described herein can be in addition to illustrating herein
Or the sequence other than the content of description is implemented.In addition, term " includes " and " having " and their any deformation, it is intended that
Cover it is non-exclusive include, for example, containing the process, method, system, product or equipment of a series of steps or units need not limit
In step or unit those of is clearly listed, but may include be not clearly listed or for these process, methods, produce
The other step or units of product or equipment inherently.
Algorithm of the invention is based on LK optical flow method, is improved on the basis of the algorithm to allow its more robust and reality
With.It is introduced below for the principle of LK optical flow algorithm.LK optical flow algorithm is based primarily upon following three hypothesis:
Brightness constancy is assumed: assuming that identical pixel in the continuous two field pictures in front and back, gray value are kept constant
Constant, formula is as follows,
F (x, t) ≡ I (x (t), t)=I (x (t+dt), t+dt)
Wherein, f (x, t) indicates the gray value of t moment pixel on the position of image coordinate x, I (x (t), t) and f (x, t)
Be it is the same, only changed a kind of expression.Dt indicates very short a period of time, and x (t+dt) indicates that the pixel x (t) of t moment is being passed through
New position after having spent the dt time, I (x (t+dt), t+dt) indicate the gray value of the original position x (t) pixel after the dt moment.
Time consistency is assumed: assuming that the frequency of Image Acquisition is very high, the time between consecutive frame is very short, then same
It the position of pixel in the next frame should be just near the position of previous frame.Its corresponding formula is as follows.
Ixu+Iyv+It=0
IxIndicate partial derivative of the image I to image x-axis, IyImage I is indicated to the partial derivative of image y-axis, u is indicated along x-axis
Speed, v indicate the speed along y-axis, ItIndicate the derivative of image at any time.
Space Consistency is assumed: assuming that other points around a pixel in very little neighborhood also all have and current point phase
Same movement, then can be used this hypothesis to construct following system, and solved using least square method.
This formula is corresponding from p1...pnThe matrix form of the equation of a point.
Simple object tracking process can be completed using above-mentioned optical flow algorithm, the present invention is changed on this basis
Into.
Fig. 1 is the flow chart of the present inventor's face tracing method.Face tracking method of the invention, especially by following steps
It realizes:
Step 1 is executed, is uniformly spread in the face frame of initial position a little, initialization needs the point set tracked, is formed initial
Frame.Common optical flow tracking needs to carry out when initial frame the angle point grid of a frame, to obtain preferable point to be tracked
For subsequent robust tracking.
Referring to Fig. 1, the present invention is proposed without angle point grid, but uniformly spread a little in the face frame of initial position,
It is subsequent that line trace is clicked through to these, it does so and is effectively reduced calculation amount.According to certain picture in Initial Face rectangle frame
Element interval selected pixels point is as initial trace point.
Step 2 is executed, optical flow tracking is carried out to all the points in face frame using optical flow method, obtains each be tracked under point
The new position at one moment.Optical flow tracking includes preceding to light stream and backward light stream, for rejecting the light stream point of tracking mistake.
It carries out optical flow tracking and optical flow field is calculated according to aforementioned three hypothesis.For the optical flow tracking of some point, it is taken
Lesser contiguous range, allow it is wherein all press the pixel that the new moment is corresponded to according to identical length and direction, further according to
Brightness constancy allows the smallest direction of its global error and length as optical flow field it is assumed that choosing.
Referring to Fig. 3, being tracked due to not extracting relatively stable angle point when initial frame, light stream is being carried out
The situation for tracking mistake occur is very easy to during tracking.Therefore, the present invention introduces on the basis of original optical flow tracking
Reverse optical flow tracking (namely forward direction light stream), i.e., original initial point set is point set regular in left-side images, by light stream
After tracking formed image right in point set, be then based on image right image to the left and tracked, obtain before to light stream with
The point set of track, then obtaining the error of initial point tracking by statistics, (namely the point set after forward direction optical flow tracking is relative to initial
The error of point set), which judge to obtain the precision currently tracked, to confirm whether current tracking is effective
Tracking.
Step 3 is executed, the position of subsequent time face frame is gone out based on the new position estimation, while further according to future position
Position can estimate the size of subsequent time face frame, obtain the variation of face scale.The trace point of all selections can get
The new position at current time, the corresponding maximum region of all the points are the region of face.
It executes step 4 and closest comparison is carried out using pixel average, variance and EOH feature, intercept the people of former frame
Face frame region and the face frame region of present frame carry out the similarity-rough set of feature, whether just to judge tracked face frame region
Really.
For the robustness for increasing tracking effect, the present invention additionally uses recently other than using optical flow method to track
Adjacent comparative approach.The position for obtaining the target that prediction obtains using optical flow algorithm first, intercepts the image of former frame target position
Region and the band of position of present frame target carry out similarity-rough set, are mainly compared using the feature in the region, using picture
Plain average value, variance and EOH feature are compared.When the higher just final confirmation tracing area of characteristic similarity is correct, below
Sketch EOH feature.
EOH feature is that feature description is carried out by establishing gradient statistic histogram in the regional area of image, special
The schematic diagram of assemble-publish code, as shown in Figure 4.
The gradient magnitude and gradient direction of each pixel in EOH feature calculation image-region, obtain a gradient map, so
Statistic histogram is divided into 9 dimensions according to gradient direction afterwards, the gradient magnitude accumulation with the same direction is got up, can be obtained most
Whole statistic histogram.Therefore gradient value in a certain direction can be obtained are as follows:
binkIndicate k-th of histogram, θ (x, y) indicates the gradient direction at image coordinate point (x, y), G ' (x, y) be
Gradient modulus value at image coordinate point (x, y).
And then the gradient value of pixel whole in regional area is got up according to above-mentioned formula accumulation, form one 9
The vector of dimension carries out arest neighbors comparison with the average value of area pixel using the vector together with variance.
The present invention proposes a kind of quick face tracking method, suitable for the numerous areas such as robot to Given Face into
The long-term tracking work of row, can be applied on service robot, operates on embedded low configuration board, it can be achieved that real-time tracking
Effect, and there is certain robustness to the variation of illumination and human face posture, there is preferable practical value.In addition to this,
Algorithm in the present invention is also relatively general track algorithm, can also be tracked to general objectives object, but need to be tracked
Target object has certain textured pattern, otherwise optical flow tracking the case where it is easy to appear error trackings.Therefore can be used for handing over
In the fields such as logical monitoring, target vehicle or specific invasion human body are tracked, to complete the work such as further processing.
In addition, the present invention correspondingly discloses a kind of quick face tracking system.Specifically, it comprises the following modules:
Face frame setup module, for uniformly spreading in the face frame of initial position a little, initialization needs the point set tracked,
Form initial frame;
Optical flow tracking module obtains each quilt for carrying out optical flow tracking to all the points in face frame using optical flow method
The new position of trace point subsequent time;Optical flow tracking module carry out before to optical flow tracking and backward optical flow tracking, for reject with
The light stream point of track mistake.Optical flow tracking module obtains new to optical flow tracking before being also used to carry out the point set of the initial position
The point set of position;To optical flow tracking after being carried out to the point set of new position;Statistics obtains the error of point set tracking.
Face frame estimation block goes out the position of subsequent time face frame based on the new position estimation, while further according to pre-
The position of measuring point can estimate the size of subsequent time face frame, obtain the variation of face scale.
Face tracking system further includes closest comparison module, for intercepting the face frame region and present frame of former frame
Face frame region carries out the similarity-rough set of feature, judges whether tracked face frame region is correct.Closest comparison module
Closest comparison is carried out using pixel average, variance and EOH feature.
It is apparent to those skilled in the art that for convenience and simplicity of description, the system of foregoing description,
The specific work process of device and unit, can refer to corresponding processes in the foregoing method embodiment, and details are not described herein.
In several embodiments provided herein, it should be understood that disclosed system, device and method can be with
It realizes by another way.For example, the apparatus embodiments described above are merely exemplary, for example, the unit
It divides, only a kind of logical function partition, there may be another division manner in actual implementation, such as multiple units or components
It can be combined or can be integrated into another system, or some features can be ignored or not executed.Another point, it is shown or
The mutual coupling, direct-coupling or communication connection discussed can be through some interfaces, the indirect coupling of device or unit
It closes or communicates to connect, can be electrical property, mechanical or other forms.
The unit as illustrated by the separation member may or may not be physically separated, aobvious as unit
The component shown may or may not be physical unit, it can and it is in one place, or may be distributed over multiple
In network unit.It can select some or all of unit therein according to the actual needs to realize the mesh of this embodiment scheme
's.
It, can also be in addition, the functional units in various embodiments of the present invention may be integrated into one processing unit
It is that each unit physically exists alone, can also be integrated in one unit with two or more units.Above-mentioned integrated list
Member both can take the form of hardware realization, can also realize in the form of software functional units.
Those of ordinary skill in the art will appreciate that all or part of the steps in the various methods of above-described embodiment is can
It is completed with instructing relevant hardware by program, which can be stored in a computer readable storage medium, storage
Medium may include: read-only memory (ROM, Read Only Memory), random access memory (RAM, Random
Access Memory), disk or CD etc..
Face tracking method provided by the present invention and system are described in detail above, for the general of this field
Technical staff, thought according to an embodiment of the present invention, there will be changes in the specific implementation manner and application range, to sum up
Described, the contents of this specification are not to be construed as limiting the invention.
Claims (10)
1. a kind of face tracking method, which comprises the following steps:
Step 1, it is uniformly spread in the face frame of initial position a little, initialization needs the point set tracked, forms initial frame;
Step 2, optical flow tracking is carried out to all the points in face frame using optical flow method, obtains each tracked point subsequent time
New position;
Step 3, the position of subsequent time face frame is gone out based on the new position estimation, while can be estimated further according to the position of future position
The size of subsequent time face frame is calculated, the variation of face scale is obtained.
2. face tracking method as described in claim 1, which is characterized in that in the step 2, before the optical flow tracking includes
To light stream and backward light stream, for rejecting the light stream point of tracking mistake.
3. face tracking method as claimed in claim 2, which is characterized in that the step 2 includes: the point of the initial position
Collect by preceding to the point set for obtaining new position after optical flow tracking;To optical flow tracking after being carried out to the point set of new position;Statistics obtains
The error of point set tracking.
4. face tracking method as described in claim 1, which is characterized in that further comprising the steps of: closest comparison is carried out,
The face frame region of the face frame region and present frame that intercept former frame carries out the similarity-rough set of feature, judges tracked people
Whether face frame region is correct.
5. face tracking method as claimed in claim 4, which is characterized in that use pixel average, variance and EOH feature
Carry out closest comparison.
6. a kind of face tracking system, which is characterized in that comprise the following modules:
Face frame setup module, for uniformly spreading in the face frame of initial position a little, initialization needs the point set tracked, is formed
Initial frame;
Optical flow tracking module obtains each be tracked for carrying out optical flow tracking to all the points in face frame using optical flow method
The new position of point subsequent time;
Face frame estimation block goes out the position of subsequent time face frame based on the new position estimation, while further according to future position
Position can estimate the size of subsequent time face frame, obtain the variation of face scale.
7. face tracking system as claimed in claim 6, which is characterized in that the optical flow tracking module carry out before to light stream with
Track and backward optical flow tracking, for rejecting the light stream point of tracking mistake.
8. face tracking system as claimed in claim 7, which is characterized in that the optical flow tracking module is also used to described
To optical flow tracking before the point set progress of initial position, the point set of new position is obtained;After being carried out to the point set of new position to light stream with
Track;Statistics obtains the error of point set tracking.
9. face tracking system as claimed in claim 6, which is characterized in that face tracking system further includes closest relatively mould
Block judges institute for intercepting the similarity-rough set of the face frame region of former frame and the face frame region progress feature of present frame
Whether the face frame region of tracking is correct.
10. face tracking system as claimed in claim 6, which is characterized in that the closest comparison module is flat using pixel
Mean value, variance and EOH feature carry out closest comparison.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710516233.0A CN109215054A (en) | 2017-06-29 | 2017-06-29 | Face tracking method and system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710516233.0A CN109215054A (en) | 2017-06-29 | 2017-06-29 | Face tracking method and system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109215054A true CN109215054A (en) | 2019-01-15 |
Family
ID=64976521
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710516233.0A Pending CN109215054A (en) | 2017-06-29 | 2017-06-29 | Face tracking method and system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109215054A (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111079670A (en) * | 2019-12-20 | 2020-04-28 | 北京百度网讯科技有限公司 | Face recognition method, face recognition device, face recognition terminal and face recognition medium |
CN115174861A (en) * | 2022-07-07 | 2022-10-11 | 广州后为科技有限公司 | Method and device for automatically tracking moving target by pan-tilt camera |
WO2023088074A1 (en) * | 2021-11-18 | 2023-05-25 | 北京眼神智能科技有限公司 | Face tracking method and apparatus, and storage medium and device |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103514441A (en) * | 2013-09-21 | 2014-01-15 | 南京信息工程大学 | Facial feature point locating tracking method based on mobile platform |
CN105469056A (en) * | 2015-11-26 | 2016-04-06 | 小米科技有限责任公司 | Face image processing method and device |
CN106296742A (en) * | 2016-08-19 | 2017-01-04 | 华侨大学 | A kind of online method for tracking target of combination Feature Points Matching |
-
2017
- 2017-06-29 CN CN201710516233.0A patent/CN109215054A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103514441A (en) * | 2013-09-21 | 2014-01-15 | 南京信息工程大学 | Facial feature point locating tracking method based on mobile platform |
CN105469056A (en) * | 2015-11-26 | 2016-04-06 | 小米科技有限责任公司 | Face image processing method and device |
CN106296742A (en) * | 2016-08-19 | 2017-01-04 | 华侨大学 | A kind of online method for tracking target of combination Feature Points Matching |
Non-Patent Citations (4)
Title |
---|
KOBI LEVI ET AL: "《Learning Object Detection from a Small Number of Examples:the Importance of Good Features》", 《CVPR’04》 * |
周千昊 等: "《基于改进EOH特征的行人检测》", 《中国图象图形学报》 * |
李泽泉: "《基于TLD框架的人脸跟踪算法的研究与实现》", 《万方数据库》 * |
许高凤 等: "《一种联合的人眼精确定位算法》", 《小型微型计算机***》 * |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111079670A (en) * | 2019-12-20 | 2020-04-28 | 北京百度网讯科技有限公司 | Face recognition method, face recognition device, face recognition terminal and face recognition medium |
CN111079670B (en) * | 2019-12-20 | 2023-11-03 | 北京百度网讯科技有限公司 | Face recognition method, device, terminal and medium |
WO2023088074A1 (en) * | 2021-11-18 | 2023-05-25 | 北京眼神智能科技有限公司 | Face tracking method and apparatus, and storage medium and device |
CN115174861A (en) * | 2022-07-07 | 2022-10-11 | 广州后为科技有限公司 | Method and device for automatically tracking moving target by pan-tilt camera |
CN115174861B (en) * | 2022-07-07 | 2023-09-22 | 广州后为科技有限公司 | Method and device for automatically tracking moving target by holder camera |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Amato et al. | Deep learning for decentralized parking lot occupancy detection | |
Feng et al. | Local background enclosure for RGB-D salient object detection | |
Roa'a et al. | Generation of high dynamic range for enhancing the panorama environment | |
CN106815842B (en) | improved super-pixel-based image saliency detection method | |
JP2008538832A (en) | Estimating 3D road layout from video sequences by tracking pedestrians | |
CN105809716B (en) | Foreground extraction method integrating superpixel and three-dimensional self-organizing background subtraction method | |
US8538079B2 (en) | Apparatus capable of detecting location of object contained in image data and detection method thereof | |
CN107341815B (en) | Violent motion detection method based on multi-view stereoscopic vision scene stream | |
US9406140B2 (en) | Method and apparatus for generating depth information | |
CN110910421A (en) | Weak and small moving object detection method based on block characterization and variable neighborhood clustering | |
CN109215054A (en) | Face tracking method and system | |
CN103714556A (en) | Moving target tracking method based on pyramid appearance model | |
CN110599522A (en) | Method for detecting and removing dynamic target in video sequence | |
TW201436552A (en) | Method and apparatus for increasing frame rate of an image stream using at least one higher frame rate image stream | |
Cuevas et al. | Moving object detection for real-time augmented reality applications in a GPGPU | |
CN114072839A (en) | Hierarchical motion representation and extraction in monocular still camera video | |
CN110909617B (en) | Living body face detection method and device based on binocular vision | |
CN109299702B (en) | Human behavior recognition method and system based on depth space-time diagram | |
CN104978558B (en) | The recognition methods of target and device | |
KR20110112143A (en) | A method for transforming 2d video to 3d video by using ldi method | |
CN107274477B (en) | Background modeling method based on three-dimensional space surface layer | |
CN104517292A (en) | Multi-camera high-density crowd partitioning method based on planar homography matrix restraint | |
CN110322479B (en) | Dual-core KCF target tracking method based on space-time significance | |
Yang et al. | Contrast limited adaptive histogram equalization for an advanced stereo visual slam system | |
Xu et al. | Moving target detection and tracking in FLIR image sequences based on thermal target modeling |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190115 |