CN112509107B - Point cloud attribute re-coloring method, device and encoder - Google Patents
Point cloud attribute re-coloring method, device and encoder Download PDFInfo
- Publication number
- CN112509107B CN112509107B CN202011396402.XA CN202011396402A CN112509107B CN 112509107 B CN112509107 B CN 112509107B CN 202011396402 A CN202011396402 A CN 202011396402A CN 112509107 B CN112509107 B CN 112509107B
- Authority
- CN
- China
- Prior art keywords
- point
- point cloud
- points
- original
- attribute
- 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.)
- Active
Links
- 238000004040 coloring Methods 0.000 title claims abstract description 43
- 238000000034 method Methods 0.000 title claims abstract description 41
- 238000012545 processing Methods 0.000 claims abstract description 19
- 238000004364 calculation method Methods 0.000 claims description 15
- 238000010276 construction Methods 0.000 claims description 3
- 238000013139 quantization Methods 0.000 description 15
- 238000010586 diagram Methods 0.000 description 4
- 238000013519 translation Methods 0.000 description 3
- 238000002310 reflectometry Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 238000007906 compression Methods 0.000 description 1
- 230000006835 compression Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T15/00—3D [Three Dimensional] image rendering
- G06T15/50—Lighting effects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
- G06T17/20—Finite element generation, e.g. wire-frame surface description, tesselation
Abstract
The invention discloses a point cloud attribute re-coloring method, a device and an encoder, wherein the point cloud attribute re-coloring method comprises the following steps: acquiring and processing an original point cloud to obtain a reconstructed point cloud; finding out the associated point corresponding to each point in the reconstructed point cloud in the original point cloud to obtain an associated point set; combining the repeated points of the original point cloud in the associated point set based on the original attribute information of all points in the associated point set to obtain the attribute information of the processed associated point set; and calculating the attribute information of each point in the reconstructed point cloud according to the processed attribute information of the associated point set. When the attribute is recoloured, the method carries out merging treatment on the repeated points in the original point cloud, reduces the influence of the repeated points on the reconstructed attribute value, and improves the quality and the coding performance of the reconstructed point cloud without increasing the coding complexity.
Description
Technical Field
The invention belongs to the technical field of point cloud coding, and particularly relates to a point cloud attribute re-coloring method, a device and an encoder.
Background
A point cloud is a set of irregularly distributed discrete points in space that represent the spatial structure and surface properties of a three-dimensional object or scene. Generally, the three-dimensional point cloud data includes geometric information representing three-dimensional space coordinates of each point, and attribute information such as colors, reflectivity and the like attached to each point, and may also have materials or other information according to application scenes.
With the continuous development of point cloud technology, compression coding of point cloud data becomes an important research problem. Currently, standards for point cloud coding are established by the national digital audio and video codec standards working group (AVS, audio Video coding Standard Workgroup of China) and the moving picture expert group (MPEG, moving Picture Experts Group) in the international organization for standardization. For the original point cloud, the geometric information and the attribute information of the point cloud are separately encoded and decoded whether the point cloud is an AVS platform or a G-PCC platform. At present, attribute coding is mainly performed on color information and reflectivity information. For existing AVS and G-PCC platforms, it is often necessary to convert the color information in the original point cloud attributes from RGB color space to luminance and chrominance color space when performing attribute encoding; the point cloud is then recoloured with the reconstructed geometric information such that the uncoded attribute information corresponds to the reconstructed geometric information, a process known as point cloud attribute recolouring. And then performs the following attribute prediction and encoding processes.
However, because the point cloud sequence itself has repeated points, when the existing AVS platform and the G-PCC platform perform geometric lossy re-coloring, the repeated points of the point cloud itself are not considered, that is, the calculation of the reconstructed attribute value will calculate the repeated points of the point cloud itself for multiple times, so as to influence the reconstructed attribute value, thereby influencing the new encoding performance.
Disclosure of Invention
In order to solve the problems in the prior art, the invention provides a point cloud attribute re-coloring method, a point cloud attribute re-coloring device and an encoder. The technical problems to be solved by the invention are realized by the following technical scheme:
a method of re-coloring point cloud attributes, comprising:
acquiring and processing an original point cloud to obtain a reconstructed point cloud;
finding out the associated point corresponding to each point in the reconstructed point cloud in the original point cloud to obtain an associated point set;
combining the repeated points of the original point cloud in the associated point set based on the original attribute information of all points in the associated point set to obtain the attribute information of the processed associated point set;
and calculating the attribute information of each point in the reconstruction point cloud according to the processed attribute information of the associated point set.
In an embodiment of the present invention, finding out a correlation point corresponding to each point in the reconstructed point cloud in the original point cloud, to obtain a correlation point set, including:
for each point in the first reconstructed point cloudFinding a plurality of points P quantized to the point in the original point cloud k (i) As a set of association points, note:
U(i)＝(P1 k (i)) k∈{1，...，D1(i)} ；
wherein D1 (i) is the number of points of U (i), P1 k (i)∈(P i ) i＝0...N-1 ，P i Representing points in the original point cloud.
In an embodiment of the present invention, based on original attribute information of all points in the association point set, merging processing is performed on repeated points of the original point cloud in the association point set to obtain attribute information of the processed association point set, including:
combining each group of repeated points of the original point cloud in the associated point set into a point respectively, and obtaining the attribute value of the combined point according to the original attribute value of each group of repeated points; wherein the attribute value of the merging point is the average value, or the median, or the mode, or the first value, or the maximum value, or the minimum value of the original attribute values of all the repeated points in the group;
and obtaining the attribute information of the processed association point set based on the original attribute value of the non-repeated point of the original point cloud in the association point set and the attribute value of the merging point.
In an embodiment of the present invention, a calculation formula of the attribute information of each point in the first reconstructed point cloud is:
wherein,representing the first reconstruction point cloud midpoint +.>Attribute value of A1 k (i) Representing the midpoint P1 of the set of associated points U (i) k (i) And point P1 k (i) Represents a point or a merging point of multiple points, w k (i) Representation point P1 k (i) Is used for the weight of the (c).
In an embodiment of the present invention, finding out a correlation point corresponding to each point in the reconstructed point cloud in the original point cloud, to obtain a correlation point set, and further including:
for each point in the second reconstructed point cloudFinding a plurality of points closest to the point in the original point cloud, and marking as:
W(i)＝(P k * (i)) k∈{1，...，C(i)} ；
wherein P is k * (i) Is associated withPoints of equal distance, C (i) is the number of points of W (i);
for each point (P i ) i＝0...N-1 Finding a point closest to the second reconstruction point cloud;
for each point in the second reconstructed point cloudThe original point is added with the +.>The set of points V (i) that are closest points are noted as a set of associated points:
V(i)＝(P2 k (i)) k∈{1，...，D2(i)} ；
wherein D2 (i) is the number of points of V (i), P2 k (i)∈(P i ) i＝0...N-1 ，P i Representing points in the original point cloud.
In an embodiment of the present invention, a calculation formula of the attribute information of each point in the second reconstruction point cloud is:
if V (i) is empty, then pointThe reconstructed attribute values of (a) are:
if V (i) is not null, then the dotThe reconstructed attribute values of (a) are:
wherein A is k * (i) Represents the original point Yun Zhongdian P k * (i) Attribute value of A2 k (i) Representing the midpoint P2 of the set of associated points V (i) k (i) And point P2 k (i) Representing a point or a combination of points.
In an embodiment of the present invention, finding out a correlation point corresponding to each point in the reconstructed point cloud in the original point cloud, to obtain a correlation point set, and further including:
for each point in the third reconstruction point cloudFinding a point P closest to the point in the original point cloud i * Wherein point P i * The corresponding attribute value is A i * ；
For each point (P i ) i＝0...N-1 Finding a point closest to the third reconstruction point cloud;
for each point in the third reconstruction point cloudThe original point is added with the +.>The set of points V' (i) that are closest points are noted as a set of associated points:
V′(i)＝(P3 k (i)) k∈{1，...，D3(i)} ；
wherein D3 (i) is the number of points of V' (i), P3 k (i)∈(P i ) i＝0...N-1 ，P i Representing points in the original point cloud.
In an embodiment of the present invention, a calculation formula of the attribute information of each point in the third reconstruction point cloud is:
if V' (i) is empty, thenThe reconstructed attribute values of (a) are:
if V' (i) is not null, thenThe reconstructed attribute values of (a) are:
wherein A3 k (i) Representing the midpoint P3 of the set of associated points V' (i) k (i) And point P3 k (i) Representing a point or a combination of points.
Another embodiment of the present invention provides a point cloud attribute re-coloring apparatus, including:
the data acquisition module is used for acquiring and processing the original point cloud to obtain a reconstructed point cloud;
the association point set construction module is used for finding out association points corresponding to each point in the reconstruction point cloud in the original point cloud to obtain an association point set;
the repeated point processing module is used for carrying out combination processing on repeated points of the original point cloud in the associated point set based on the original attribute information of all points in the associated point set so as to obtain the attribute information of the processed associated point set;
and the attribute calculation module is used for calculating the attribute information of each point in the reconstruction point cloud according to the processed attribute information of the associated point set.
An embodiment of the present invention provides a point cloud encoder, which includes a geometry encoding system and an attribute encoding system, where the attribute encoding system includes a point cloud attribute re-coloring device as described in the above embodiment.
Compared with the prior art, the invention has the beneficial effects that:
according to the point cloud attribute re-coloring method provided by the invention, the influence of the repeated points of the geometric quantization point cloud on the re-attribute value is considered, when attribute re-coloring is carried out, the repeated points in the original point cloud are combined, the influence of the repeated points on the re-attribute value is reduced, and the attribute re-forming performance is improved while the coding complexity is not increased.
Drawings
Fig. 1 is a schematic flow chart of a point cloud attribute re-coloring method according to an embodiment of the present invention;
fig. 2 is a schematic structural diagram of a point cloud attribute re-coloring device according to an embodiment of the present invention;
fig. 3 is a schematic structural diagram of a point cloud encoder according to an embodiment of the present invention.
Detailed Description
The present invention will be described in further detail with reference to specific examples, but embodiments of the present invention are not limited thereto.
Example 1
Referring to fig. 1, fig. 1 is a flow chart of a point cloud attribute re-coloring method according to an embodiment of the present invention, including:
acquiring and processing an original point cloud to obtain a reconstructed point cloud;
finding out the associated point corresponding to each point in the reconstructed point cloud in the original point cloud to obtain an associated point set;
combining the repeated points of the original point cloud in the associated point set based on the original attribute information of all points in the associated point set to obtain the attribute information of the processed associated point set;
and calculating the attribute information of each point in the reconstruction point cloud according to the processed attribute information of the associated point set.
In practical applications, the reconstructed point cloud may be obtained in various manners, for example, in an AVS platform, the reconstructed point cloud may be obtained by coordinate translation quantization, then subjected to a re-coloring process, or after geometric reconstruction, reconstructed point cloud information may be obtained, and then subjected to quantization process. In the G-PCC platform, quantization is also performed based on point cloud data obtained after geometric reconstruction.
Aiming at the three reconstructed point clouds and comprehensively considering the influence of the repeated points of the geometric quantization point clouds on the reconstructed attribute values, the embodiment provides a point cloud attribute re-coloring method based on repeated point combination. When the method provided by the embodiment is used for attribute re-coloring, the repeated points in the original point cloud are combined, so that the influence of the repeated points on the reconstructed attribute value is reduced, and the quality and the coding performance of the reconstructed point cloud are improved while the coding complexity is not increased.
Example two
The method for re-coloring the point cloud attribute provided in the first embodiment is described in detail below by specifically using the AVS platform.
In the current AVS platform, there are two methods of re-coloring under geometrically lossy conditions. One is a fast re-coloring method, which is performed after the coordinate translation quantization. In this embodiment, the reconstructed point cloud obtained after the coordinate translation quantization is referred to as a first reconstructed point cloud. The method specifically comprises the following steps:
step one: and acquiring and processing the original point cloud to obtain a first reconstruction point cloud.
Specifically, in the frame of the point cloud AVS encoder, first, the geometric information of the original point cloud data is subjected to coordinate transformation, so that the point cloud is all contained in one bounding box. And then quantifying to obtain a first reconstruction point cloud. This quantization mainly plays a scaling role, and since the quantization is rounded, the geometric information of a part of points is the same, and in particular, whether to remove the repeated points can be determined according to parameters.
Let the geometrical information of the original point cloud and the first reconstruction point cloud be represented as (P) i ) i＝0...N-1 Andwherein N and N rec Points in the original point cloud and the first reconstructed point cloud are represented, respectively. If the repetition point generated by quantization is removed, N rec ＜N。
Step two: and finding out the associated point corresponding to each point in the first reconstruction point cloud from the original point cloud to obtain an associated point set.
For the rapid re-coloring method, as the point cloud sequence has repeated points, only the repeated points obtained by quantization are considered when the geometrical damage is used for re-coloring, and the repeated points of the point cloud are not considered, namely, the repeated points of the point cloud are calculated for a plurality of times by calculating the reconstruction attribute value, so that the reconstruction attribute value is influenced. For example: points a and B are quantized to point C, where point B has tens of repeated points, and the calculated attribute value of point C is affected by these repeated points, so that the result is closer to the attribute value of point B.
Therefore, in the embodiment, when fast re-coloring is performed, the merging process is performed on the repeated points in the original point cloud, so as to reduce the influence of the repeated points on the attribute calculation.
Specifically, the second step includes:
for each point in the first reconstructed point cloudFinding a plurality of points quantized to the point in the original point cloud, and marking the points as a set of associated points as:
U(i)＝(P1 k (i)) k∈{1，...，D1(i)} ；
wherein D1 (i) is the number of points of U (i), P1 k (i)∈(P i ) i＝0...N-1 ，P i Representing points in the original point cloud.
Step three: and merging repeated points of the original point cloud in the associated point set based on the original attribute information of all points in the associated point set to obtain the attribute information of the processed associated point set.
First, each point in the associated point set obtained in the second step carries its original attribute information, but there is a duplicate point of the original point cloud in the point set. Therefore, these repetition points need to be processed.
Specifically, each group of repeated points of the original point cloud in the associated point set are respectively combined into a point, and the attribute value of the combined point is obtained according to the original attribute value of each group of repeated points.
More specifically, the average, or median, or mode, or first value, or maximum, or minimum of the original attribute values of all the repeated points of the set may be taken as the attribute value of the merge point.
Preferably, the original attribute mean of the set of repeated points may be taken as the attribute value of the merge point.
Then, attribute information of the associated point set is obtained based on the original attribute values of the non-duplicate points of the original point cloud in the associated point set and the attribute values of the merge points.
Step four: and calculating the attribute information of each point in the first reconstruction point cloud according to the processed attribute information of the associated point set.
In this embodiment, a calculation formula of the attribute information of each point in the first reconstruction point cloud is:
wherein,representing the first reconstruction point cloud midpoint +.>Attribute value of A1 k (i) Representing the midpoint P1 of the set of associated points U (i) k (i) And point P1 k (i) Represents a point or a merging point of multiple points, w k (i) Representation point P1 k (i) Is used for the weight of the (c).
Further, w k (i) The calculation formula of (2) is as follows:
wherein,representing the point of concentration P1 of the associated point k (i) The value e divided by the quantization step is a constant, which takes 0.1.
And obtaining the attribute information of all points in the first reconstruction point cloud, and finishing the point cloud attribute re-coloring processing.
Example III
The embodiment provides another point cloud re-coloring method of the AVS platform, which is performed after geometric reconstruction, and the point cloud data after geometric reconstruction of the AVS platform is called as second reconstruction point cloud.
The method specifically comprises the following steps:
step 1: and acquiring and processing the original point cloud to obtain a second reconstruction point cloud.
Specifically, the original point cloud and the second reconstructed point cloud (point cloud after geometric reconstruction) are respectively represented as (P) i ) i＝0...N-1 Andwherein N and N rec Points in the original point cloud and the second reconstructed point cloud are represented, respectively. If the repeated points generated by geometric quantization in the geometric reconstruction process are removed, N rec ＜N。
Step 2: and finding out the associated point corresponding to each point in the second reconstruction point cloud from the original point cloud to obtain an associated point set.
The embodiment mainly obtains the associated point set corresponding to each point in the second reconstruction point cloud by using a nearest neighbor search method, and is specifically realized through a KTree data structure.
Specifically, step 2 includes:
21 For each point in the second reconstructed point cloudFinding a plurality of points closest to the point in the original point cloud, and marking as:
W(i)＝(P k * (i)) k∈{1，...，C(i)} ；
wherein P is k * (i) Is associated withPoints of equal distance, C (i) is the number of points of W (i); w (i) may comprise one point or a plurality of points.
22 For each point (P) in the original point cloud i ) i＝0...N-1 Finding the nearest point from the reconstructed point cloud;
23 For each point in the reconstructed point cloudThe original point is added with the +.>The set of points V (i) that are closest points are noted as a set of associated points:
V(i)＝(P2 k (i)) k∈{1，...，D2(i)} ；
wherein D2 (i) is the number of points of V (i), P2 k (i)∈(P i ) i＝0...N-1 ，P i Representing points in the original point cloud, V (i) may be null or may include one or more points.
Step 3: and merging repeated points of the original point cloud in the associated point set based on the original attribute information of all points in the associated point set to obtain the attribute information of the processed associated point set.
As in the first embodiment, each set of all the repeated points may be combined into one point, and the original attribute mean, or median, or mode of all the repeated points in the set may be used as the attribute value of the combined point, and the original attribute value of the non-repeated points may be reserved, so as to obtain the attribute information of all the points in the associated point set.
Step 4: and calculating the attribute information of each point in the second reconstruction point cloud according to the processed attribute information of the associated point set.
Specifically, a calculation formula of the attribute information of each point in the second reconstruction point cloud is:
if V (i) is empty, then pointThe reconstructed attribute values of (a) are:
if V (i) is not null, then the dotThe reconstructed attribute values of (a) are:
wherein A is k * (i) Represents the original point Yun Zhongdian P k * (i) Attribute value of A2 k (i) Representing the midpoint P2 of the set of associated points V (i) k (i) And point P2 k (i) Representing a point or a combination of points.
And obtaining the attribute information of all points in the second reconstruction point cloud, and finishing the point cloud attribute re-coloring processing.
Example III
The embodiment describes the point cloud attribute re-coloring method provided in the first embodiment in detail through a G-PCC platform.
In current MPEG G-PCC, under geometrically lossy conditions, a re-coloring after geometrical reconstruction is required, wherein computation involving nearest neighbor lookups also uses KDTree data structures. In this embodiment, the point cloud data after geometric reconstruction of the G-PCC platform is referred to as a third reconstruction point cloud.
Step A: and obtaining and processing the original point cloud to obtain a third reconstruction point cloud.
Specifically, let the original point cloud and the third reconstructed point cloud (point cloud after geometric reconstruction) geometric information be represented as (P) i ) i＝0...N-1 Andwherein N and N rec Points in the original point cloud and the third reconstruction point cloud are represented, respectively. If the repeated points generated by geometric quantization in the geometric reconstruction process are removed, N rec ＜N。
And (B) step (B): finding out an associated point corresponding to each point in the third reconstruction point cloud in the original point cloud to obtain an associated point set, wherein the method specifically comprises the following steps:
b1 For each point in the third reconstruction point cloudFinding the nearest point P from the original point cloud i * Wherein point P i * The corresponding attribute value is A i * ；
B2 For each point (P) in the original point cloud i ) i＝0...N-1 Finding the nearest point from the reconstructed point cloud;
b3 For each point in the reconstructed point cloudThe original point is added with the +.>The set of points V' (i) that are closest points are noted as a set of associated points:
V′(i)＝(P3 k (i)) k∈{1，...，D3(i)} ；
wherein D3 (i) is the number of points of V' (i), P3 k (i)∈(P i ) i＝0...N-1 ，P i Representing points in the original point cloud, V' (i) may be null or may include one or more points.
Step C: and merging repeated points of the original point cloud in the associated point set based on the original attribute information of all points in the associated point set to obtain the processed attribute information of the associated point set.
As in the first and second embodiments, each set of all the repeated points may be combined into one point, and the original attribute mean, or median, or mode of all the repeated points in the set may be used as the attribute value of the combined point, and the original attribute value of the non-repeated points may be reserved, so as to obtain the attribute information of all the points in the associated point set.
Step D: and calculating the attribute information of each point in the third reconstruction point cloud according to the processed attribute information of the associated point set.
Specifically, the calculation formula of the attribute information of each point in the third reconstruction point cloud is:
if V' (i) is empty, thenThe reconstructed attribute values of (a) are:
if V' (i) is not null, thenThe reconstructed attribute values of (a) are:
wherein A3 k (i) Representing the midpoint P3 of the set of associated points V' (i) k (i) And point P3 k (i) Representing a point or a combination of points.
And obtaining the attribute information of all points in the third reconstruction point cloud, and finishing the point cloud attribute re-coloring processing.
Example IV
On the basis of the first embodiment, the present embodiment provides a point cloud attribute re-coloring device, please refer to fig. 2, fig. 2 is a schematic structural diagram of the point cloud attribute re-coloring device provided in the embodiment of the present invention, which includes:
the data acquisition module 11 is used for acquiring and processing the original point cloud to obtain a reconstructed point cloud;
the association point set construction module 12 is configured to find an association point corresponding to each point in the reconstructed point cloud in the original point cloud, so as to obtain an association point set;
the repeated point processing module 13 is configured to combine the repeated points of the original point cloud in the associated point set based on the original attribute information of all points in the associated point set, so as to obtain the attribute information of the processed associated point set;
and the attribute calculation module 14 is used for calculating the attribute information of each point in the reconstruction point cloud according to the processed attribute information of the associated point set.
The point cloud attribute re-coloring device provided in this embodiment may implement the point cloud attribute re-coloring method described in the first embodiment, which may be applied to the AVS platform and the G-PCC platform, and the specific implementation process is referred to in the second embodiment to the fourth embodiment, which is not described herein again.
Example five
On the basis of the fourth embodiment, the present embodiment provides a point cloud encoder, please refer to fig. 3, fig. 3 is a schematic structural diagram of a point cloud reconstruction system provided by the embodiment of the present invention, which includes a geometric coding system and an attribute coding system, where the attribute coding system includes the point cloud attribute re-coloring device described in the fourth embodiment.
When the point cloud encoder provided by the embodiment performs attribute encoding, the influence of the repeated points of the geometric quantization point cloud on the reconstructed attribute value is considered, the repeated points of the point cloud are combined in the re-coloring process, the influence of the repeated points on the reconstructed attribute value is reduced, and the quality and the encoding performance of the reconstructed point cloud are improved while the encoding complexity is not increased.
The foregoing is a further detailed description of the invention in connection with the preferred embodiments, and it is not intended that the invention be limited to the specific embodiments described. It will be apparent to those skilled in the art that several simple deductions or substitutions may be made without departing from the spirit of the invention, and these should be considered to be within the scope of the invention.
Claims (10)
1. A method for re-coloring point cloud attributes, comprising:
acquiring and processing an original point cloud to obtain a reconstructed point cloud;
finding out the associated point corresponding to each point in the reconstructed point cloud in the original point cloud to obtain an associated point set;
combining the repeated points of the original point cloud in the associated point set based on the original attribute information of all points in the associated point set to obtain the attribute information of the processed associated point set;
and calculating the attribute information of each point in the reconstruction point cloud according to the processed attribute information of the associated point set.
2. The method of claim 1, wherein finding out associated points in the original point cloud that correspond to each point in the reconstructed point cloud to obtain an associated point set, comprises:
for each point in the first reconstructed point cloudFinding a plurality of points P quantized to the point in the original point cloud k (i) As a set of association points, note:
U(i)＝(P1 k (i)) k∈{1，...，D1(i)} ；
wherein D1 (i) is the number of points of U (i), P1 k (i)∈(P i ) i＝0...N-1 ，P i Representing points in the original point cloud.
3. The method of claim 1, wherein merging the repeated points of the original point cloud in the associated point set based on the original attribute information of all points in the associated point set to obtain the attribute information of the processed associated point set, includes:
combining each group of repeated points of the original point cloud in the associated point set into a point respectively, and obtaining the attribute value of the combined point according to the original attribute value of each group of repeated points; wherein the attribute value of the merging point is the average value, or the median, or the mode, or the first value, or the maximum value, or the minimum value of the original attribute values of all the repeated points in the group;
and obtaining the attribute information of the processed association point set based on the original attribute value of the non-repeated point of the original point cloud in the association point set and the attribute value of the merging point.
4. The method for re-coloring the attribute of the point cloud according to claim 2, wherein the calculation formula of the attribute information of each point in the first reconstructed point cloud is:
wherein,representing the first reconstruction point cloud midpoint +.>Attribute value of A1 k (i) Representing the midpoint P1 of the set of associated points U (i) k (i) And point P1 k (i) Represents a point or a merging point of multiple points, w k (i) Representing point P1 k (i) Is used for the weight of the (c).
5. The method of claim 1, wherein finding out associated points in the original point cloud that correspond to each point in the reconstructed point cloud, and obtaining an associated point set, further comprises:
for each point in the second reconstructed point cloudFinding a plurality of points closest to the point in the original point cloud, and marking as:
W(i)＝(P k * (i)) k∈{1，...，C(i)} ；
wherein P is k * (i) Is associated withPoints of equal distance, C (i) is the number of points of W (i);
for each point (P i ) i＝0...N-1 Finding a point closest to the second reconstruction point cloud;
for each point in the second reconstructed point cloudThe original point is added with the +.>The set of points V (i) that are closest points are noted as a set of associated points:
V(i)＝(P2 k (i)) k∈{1，...，D2(i)} ；
wherein D2 (i) is the number of points of V (i), P2 k (i)∈(P i ) i＝0...N-1 ，P i Representing points in the original point cloud.
6. The method for re-coloring the attribute of the point cloud according to claim 5, wherein the calculation formula of the attribute information of each point in the second reconstructed point cloud is:
if V (i) is empty, then pointThe reconstructed attribute values of (a) are:
if V (i) is not null, then the dotThe reconstructed attribute values of (a) are:
wherein A is k * (i) Represents the original point Yun Zhongdian P k * (i) Attribute value of A2 k (i) Representing the midpoint P2 of the set of associated points V (i) k (i) And point P2 k (i) Representing a point or a combination of points.
7. The method of claim 1, wherein finding out associated points in the original point cloud that correspond to each point in the reconstructed point cloud, and obtaining an associated point set, further comprises:
for each point in the third reconstruction point cloudFinding a point P closest to the point in the original point cloud i * Wherein point P i * The corresponding attribute value is A i * ；
For each point (P i ) i＝0...N-1 Finding a third reconstruction point cloud from the first reconstruction point cloudThe closest point;
for each point in the third reconstruction point cloudThe original point is added with the +.>The set of points V' (i) that are closest points are noted as a set of associated points:
V′(i)＝(P3 k (i)) k∈{1，...，D3(i)} ；
wherein D3 (i) is the number of points of V' (i), P3 k (i)∈(P i ) i＝0...N-1 ，P i Representing points in the original point cloud.
8. The method for re-coloring the attributes of the point cloud as claimed in claim 7, wherein the calculation formula of the attribute information of each point in the third reconstruction point cloud is:
if V' (i) is empty, thenThe reconstructed attribute values of (a) are:
if V' (i) is not null, thenThe reconstructed attribute values of (a) are:
wherein A3 k (i) Representing the midpoint P3 of the set of associated points V' (i) k (i) And point P3 k (i) Representing a point or a combination of points.
9. A point cloud attribute re-coloring apparatus, comprising:
the data acquisition module (11) is used for acquiring and processing the original point cloud to obtain a reconstructed point cloud;
the association point set construction module (12) is used for finding out association points corresponding to each point in the reconstruction point cloud in the original point cloud to obtain an association point set;
the repeated point processing module (13) is used for carrying out combination processing on repeated points of the original point cloud in the associated point set based on the original attribute information of all points in the associated point set so as to obtain the attribute information of the processed associated point set;
and the attribute calculation module (14) is used for calculating the attribute information of each point in the reconstruction point cloud according to the processed attribute information of the associated point set.
10. A point cloud encoder comprising a geometry encoding system and an attribute encoding system, wherein the attribute encoding system comprises the point cloud attribute re-coloring apparatus of claim 9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011396402.XA CN112509107B (en) | 2020-12-03 | 2020-12-03 | Point cloud attribute re-coloring method, device and encoder |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011396402.XA CN112509107B (en) | 2020-12-03 | 2020-12-03 | Point cloud attribute re-coloring method, device and encoder |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112509107A CN112509107A (en) | 2021-03-16 |
| CN112509107B true CN112509107B (en) | 2024-02-20 |
Family
ID=74969559
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011396402.XA Active CN112509107B (en) | 2020-12-03 | 2020-12-03 | Point cloud attribute re-coloring method, device and encoder |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112509107B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113179410B (en) * | 2021-06-10 | 2022-08-23 | 上海交通大学 | Point cloud attribute coding and decoding method, device and system |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109257604A (en) * | 2018-11-20 | 2019-01-22 | 山东大学 | A kind of color attribute coding method based on TMC3 point cloud encoder |
| CN110796724A (en) * | 2018-07-31 | 2020-02-14 | 英特尔公司 | Point cloud viewpoint and scalable compression/decompression |
| WO2020197086A1 (en) * | 2019-03-25 | 2020-10-01 | 엘지전자 주식회사 | Point cloud data transmission device, point cloud data transmission method, point cloud data reception device, and point cloud data reception method |
| CN111953998A (en) * | 2020-08-16 | 2020-11-17 | 西安电子科技大学 | Point cloud attribute coding and decoding method, device and system based on DCT (discrete cosine transformation) |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11113845B2 (en) * | 2017-09-18 | 2021-09-07 | Apple Inc. | Point cloud compression using non-cubic projections and masks |
| US10904564B2 (en) * | 2018-07-10 | 2021-01-26 | Tencent America LLC | Method and apparatus for video coding |
-
2020
- 2020-12-03 CN CN202011396402.XA patent/CN112509107B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110796724A (en) * | 2018-07-31 | 2020-02-14 | 英特尔公司 | Point cloud viewpoint and scalable compression/decompression |
| CN109257604A (en) * | 2018-11-20 | 2019-01-22 | 山东大学 | A kind of color attribute coding method based on TMC3 point cloud encoder |
| WO2020197086A1 (en) * | 2019-03-25 | 2020-10-01 | 엘지전자 주식회사 | Point cloud data transmission device, point cloud data transmission method, point cloud data reception device, and point cloud data reception method |
| CN111953998A (en) * | 2020-08-16 | 2020-11-17 | 西安电子科技大学 | Point cloud attribute coding and decoding method, device and system based on DCT (discrete cosine transformation) |
Non-Patent Citations (2)
| Title |
|---|
| 基于IFC标准的建筑构件点云信息处理方法;徐照;康蕊;孙宁;;东南大学学报(自然科学版)(06);全文 * |
| 基于物体逻辑状态推理的未知物体视觉分割;包加桐;宋爱国;洪泽;沈天鹤;唐鸿儒;;机器人(04);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112509107A (en) | 2021-03-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10455251B2 (en) | Methods and apparatuses for coding and decoding depth map | |
| CN1270543C (en) | Method and device for computing coding dynamic images at fixed complicacy | |
| US5699121A (en) | Method and apparatus for compression of low bit rate video signals | |
| CN108495135B (en) | Quick coding method for screen content video coding | |
| CN111953998A (en) | Point cloud attribute coding and decoding method, device and system based on DCT (discrete cosine transformation) | |
| CN106170093B (en) | Intra-frame prediction performance improving coding method | |
| EP2723071A1 (en) | Encoder, decoder and method | |
| US20200304792A1 (en) | Quantization step parameter for point cloud compression | |
| US9245353B2 (en) | Encoder, decoder and method | |
| JP2008527809A (en) | Process for image compression and decompression acceleration | |
| CN112509107B (en) | Point cloud attribute re-coloring method, device and encoder | |
| US20190114805A1 (en) | Palette coding for color compression of point clouds | |
| US5881183A (en) | Method and device for encoding object contour by using centroid | |
| Wei et al. | Enhanced intra prediction scheme in point cloud attribute compression | |
| CN104902256B (en) | A kind of binocular stereo image decoding method based on motion compensation | |
| JPS63215281A (en) | Picture signal transmitter | |
| CN116325732A (en) | Decoding and encoding method, decoder, encoder and encoding and decoding system of point cloud | |
| WO2022217472A1 (en) | Point cloud encoding and decoding methods, encoder, decoder, and computer readable storage medium | |
| US20240029202A1 (en) | Super-resolution video processing method and system for effective video compression | |
| WO2020066307A1 (en) | Image decoding device, image encoding device, image processing system, and program | |
| WO2023023918A1 (en) | Decoding method, encoding method, decoder and encoder | |
| US20240040101A1 (en) | Method and device for compressing data representative of a volumetric three-dimensional scene with a view to real-time decompression for online viewing | |
| WO2022140937A1 (en) | Point cloud encoding method and system, point cloud decoding method and system, point cloud encoder, and point cloud decoder | |
| WO2023123284A1 (en) | Decoding method, encoding method, decoder, encoder, and storage medium | |
| Perera et al. | Evaluation of compression schemes for wide area video |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |