Video compression stream

Abstract

To provide a video compression stream from a video camera adapted for capturing a scene.SOLUTION: A method performed by a background blurring system identifies a first pixel included in a background at a first point in time in a first image. The method further determines a blurred pixel value of the first pixel. Subsequently, the method provides a blurred pixel value in a video compression stream at the first point in time. The method furthermore identifies in a second image of a scene, at a subsequent second point in time that the first pixel has changed to be included in a foreground of the second image. Moreover, the method provides a non-blurred pixel value, identified at the second point in time, in the video compression stream. The method further identifies at a third point in time that the first pixel has changed to be included in a background of a third image. The first pixel has remained unchanged from an intermediate point in time to the third point in time. Furthermore, when the third point in time is within an intermediate background merge time period from the intermediate point in time, the method provides the blurred pixel value in the video compression stream.SELECTED DRAWING: Figure 3

Description

Technical Field

[0001] The present disclosure relates to providing a video compression stream from a video camera adapted to capture a scene.

Background Art

[0002] For example, by using a video camera for surveillance purposes, it becomes possible to see what is happening in a specific environment such as an office, industrial land, part of a city, square, block, road, etc., on-site and / or at a later opportunity. The surrounding video stream, i.e., the scene, may help to provide insights and information related to the scene that are revealed and / or potentially interesting in some way. However, video streams such as live video streams from video cameras are generally costly and require a higher bitrate. Therefore, it is known to compress such video streams using various known techniques. Furthermore, it is known to blur the background portion to make it less distinct, so that one or more persons and / or objects in the foreground can remain in focus, while the background may appear out of focus. However, while blurring the background can potentially be efficient with respect to privacy issues and may lead to a reduction in bitrate consumption, it is still difficult to reduce the computational cost and / or bitrate in the video stream.

[0003] Accordingly, an object of embodiments herein is to provide an approach for providing a video compression stream, e.g., from a video camera adapted to capture a scene, in an improved and / or alternative way.

[0004] The above object can be achieved by the subject matter disclosed herein. Embodiments are described in the appended claims, the following description and the drawings.

Summary of the Invention

Problems to be Solved by the Invention

[0005] The disclosed subject matter relates to a method executed by a background blurring system to provide a video compression stream from a video camera adapted to capture a scene. The background blurring system identifies at least a first pixel included in the background of a first image of a scene, at a first point in time, within the first image of the scene. The background blurring system further determines a blurred pixel value for at least the first pixel. Further, the background blurring system provides the blurred pixel value into a video compression stream, subsequently at the first point in time. The background blurring system identifies that at least the first pixel has changed such that it is configured in the foreground of a second image, in an image of a second scene, subsequently at a second point in time. Further, the background blurring system provides, into the video compression stream, subsequently at the second point in time, continuously and / or intermittently, a non-blurred pixel value identified for at least the first pixel. The background blurring system further identifies that at least the first pixel has changed such that it is included in the background of a third image of a scene, in the third image of the scene, subsequently at a third point in time, where it is identified that at least the first pixel value remains unchanged from an intermediate point in time up to the third point in time. Further, when the third point in time is within a pre-determinable background merge time limit from the intermediate point in time, the background blurring system provides, subsequently at the third point in time, the blurred pixel value into the video compression stream.

[0006] The disclosed subject matter further relates to a background blur system for providing a video compression stream from a video camera adapted to capture a scene. The background blur system includes a first identification unit that identifies at least a first pixel included in the background of a first image of the scene at a first point in time, and / or adapts to. The background blur system further includes a first blurring unit that determines a blurred pixel value for at least the first pixel, and / or adapts to. Further, the background blur system includes a first providing unit that subsequently provides the blurred pixel value in the video compression stream at the first point in time, and / or adapts to. The background blur system further includes a second identification unit that subsequently identifies at a second point in time that at least the first pixel has changed to be included in the foreground of a second image of the scene, and / or adapts to. Further, the background blur system includes a second providing unit that subsequently provides, continuously and / or intermittently, the non-blurred pixel value identified for at least the first pixel in the video compression stream at the second point in time, and / or adapts to. The background blur system further includes a third identification unit that subsequently identifies at a third point in time that at least the first pixel has changed to be configured in the background of a third image of the scene, and / or adapts to, where at least the first pixel value is unchanged from an intermediate point in time to the third point in time. Further, the background blur system includes a third providing unit that subsequently provides the blurred pixel value in the video compression stream at the third point in time when the third point in time is within a pre-determinable background merge time limit from the intermediate point in time, and / or adapts to.

[0007] Further, the disclosed subject matter relates to a video camera comprising the background blur system described herein.

[0008] Furthermore, the disclosed subject matter relates to a computer program product including a computer-readable medium, or computer program code means stored in a processor, or a computer program including a processor, configured to cause steps of the disclosed background blurring system to be executed.

[0009] The disclosed subject matter further relates to a non-transitory computer-readable storage medium stored in the computer program product.

[0010] Therefore, an approach is disclosed that enables high computational efficiency and / or high bitrate efficiency background blur. That is, in a first image of a scene captured by a video camera, at a first time point, at least a first pixel configured in the background of the first image is identified. Thus, for example, as a result of the scene being composed of stationary and / or immobile people and / or objects, at said time point, one or more pixels that are considered to belong to the background area of the image of the captured scene are determined. Further, that is, since a blurred pixel value for at least the first pixel is determined, a blur value is also set for at least the first identified background pixel, and as a result, the pixel may be recognized as blurred, out of focus, and / or unclear. Further, that is, subsequently, the blurred pixel value is provided into the video compression stream at the first time point, so that the determined blurred pixel value for at least the first pixel of the first captured image is utilized for the video compression stream. Accordingly, in the video compression stream, by utilizing blurred pixels instead of non-blurred pixel values for one or more identified background pixel values, the bitrate can be reduced and / or privacy issues can be solved. Further, at that time, the background pixels may be blurred in contrast to the non-blurred foreground pixels, and objects including the foreground pixels, such as moving people and / or objects, may be in focus and / or highlighted compared to the immobile people and / or objects including the background pixels. Further, in a second image of the scene, at a subsequent second time point, it is identified that at least the first pixel has changed to be included in the foreground of the second image. Thus, in the second image of the scene occurring at any second time point, one or more pixels previously identified as belonging to the background are determined to have changed to belong to the foreground area of the second captured image, for example, as a result of one or more people and / or objects entering the scene.Furthermore, subsequently, the identified non-blurred pixel values of at least the first pixel at the second time point are provided - continuously and / or intermittently - in the video compression stream, so that the identified pixel values of at least the first pixel related to the second captured image, - and their identified changing pixel values in any number of subsequent captured images of the scene - are utilized for the video compression stream. Thus, for one or more identified foreground pixels within the video compression stream, by utilizing the non-blurred pixel values that are updated continuously and / or intermittently, foreground pixels such as, for example, one or more moving persons, and / or objects can be focused compared to the blurred background pixels. Additionally, in the third image of the scene, subsequently at the third time point, it is identified that at least the first pixel has changed such that it is included in the background of the third image, so that since at least the first pixel value remains unchanged from the intermediate time point to the third time point, in the third image of the scene occurring at any third time point, one or more pixels that were previously identified as belonging to the background and subsequently identified as belonging to the foreground, as a result of, for example, one or more persons and / or objects leaving the scene, and / or being at least partially stationary, are determined to have changed again to belong to the background region after remaining unchanged after the intermediate time point. That is, subsequently when the third time point is - when the third time point is within a pre-determinable background merge time limit from the intermediate time point - the blurred pixel values are provided for the video compression stream following the third time point, and if the time range from the intermediate time point to - that is, at least the first pixel when it last changed that pixel - the third time point, - that is, at least the first pixel changes from a foreground pixel to a background pixel again - is within the set time, then the previously determined blurred pixel values of at least the first pixel are used again in the video compression stream.That is, when at least a first pixel changes from an invariant foreground pixel to a background pixel again within a pre-determinable time period, for example, within the background merge time point, as exemplified by one or more persons and / or objects passing through the scene, for example, set to 5 seconds and / or not stationary beyond the background merge time limit, the blurred pixel that has already been determined - when subsequently determined at a first time point, that is, when at least the first pixel was previously identified as a background pixel - is reused in the video compression stream. Therefore, by reusing the already determined blurred pixel values, it is considered that it is also effective when the background merge time limit criterion is met, and it is considered unnecessary to determine the updated blurred pixel values of at least the first pixel, and accordingly, a reduced computational cost and / or efficient bitrate compression of the video compression stream are disclosed.

[0011] Therefore, an improved and / or alternative method for providing a video compression stream, such as a video compression stream, is introduced from a video camera adapted to capture a scene.

[0012] The technical features and corresponding advantages of the above method are further described in detail below.

[0013] Various aspects of non-limiting embodiments, including specific features and advantages, will be readily understood from the following detailed description and the accompanying drawings.

Brief Description of the Drawings

[0014]

Figure 1

Figure 2

Figure 3

Best Mode for Carrying Out the Invention

[0015] Hereinafter, non-limiting embodiments of the present disclosure will be described more fully with reference to the accompanying drawings, in which presently preferred embodiments of the present disclosure are shown. However, the present disclosure may be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. Like reference numerals refer to like elements throughout. The dashed lines of some of the boxes in the figures indicate that these units or operations are optional and not essential.

[0016] Hereinafter, an approach is disclosed that enables computationally efficient and / or bitrate-efficient background blur according to embodiments of the present specification related to providing a video compression stream from a video camera adapted to capture a scene.

[0017] Referring now to the drawings, and in particular FIGS. 1-2, FIGS. 1a-e show schematic diagrams of exemplary video compression stream images provided by an exemplary background blur system 1 according to embodiments of the present disclosure, and FIG. 2 further shows a schematic block diagram of an exemplary background blur system 1 according to embodiments of the present disclosure. The background blur system 1 may be adapted and / or configured to provide a video compression stream from a video camera 2 suitable for capturing a scene.

[0018] The video camera 2 may refer to any video camera suitable for taking images, for example a known one. According to one example, the video camera 2 may further be included as part of an exemplary surveillance system (not shown), a video camera arrangement with additional components, and / or a network (not shown). The scenes suitable for the video camera 2 to photograph can be represented by any scenes such as offices, industrial areas, parts of cities, open squares, blocks, roads, etc. The said scene may further optionally be a scene to be monitored. Furthermore, the video camera 2 is arranged to be relatively close to one or more potential persons and / or objects in the scene, and the said objects and / or persons may constitute a relatively large part of the captured image of the scene. The background blur system 1 may be included in the video camera 2, or alternatively, the background blur system 1 may be provided at least partially outside the video camera 2, such as in an exemplary video camera arrangement or surveillance system, and / or one or more servers. Furthermore, the video camera 2 and / or the background blur system 1 may be equipped with components that support pixel blur. The video compression stream to be provided can be of any feasible size and / or length, and furthermore, considering the constraints of the features disclosed herein, it can be of any feasible format. The term "background blur system" may also refer to a "processing system", a "compression system", and / or a "video stream background blur system", while "for the provision of a video compression stream" may also refer to "for the provision, generation, occurrence, display, storage, and / or preservation of a video compression stream", and may also refer to "for the background blur processing of a video stream" and / or "for the processing of a video stream". On the other hand, the term "video compression stream" may refer to a "compressed video stream" and / or a "compressed video signal" throughout the present disclosure, and according to one example, may further refer to a "video compression live stream".Furthermore, the "video camera adapted to capture a scene" may refer to the "video camera adapted to capture the surroundings". According to one example, it may refer to the "surveillance video camera adapted to capture a scene", and / or the "video camera adapted to capture a scene for surveillance".

[0019] The background blur system 1 may be adapted and / or configured, for example, by means of the first identification unit 101, to identify at least a first pixel included in the background of the first image within the first image of the scene at a first point in time. Thereby, at that point in time, one or more pixels that are considered to belong to the background area of the captured image of the scene are determined. According to one example, the one or more pixels considered to be background pixels are due to, for example, a scene including static and / or non-moving persons and / or objects. For example, it should be noted that the persons and / or objects constituting the background pixels do not necessarily have to be arranged further away from the video camera 2 than the persons and / or objects constituting the foreground pixels, and they may be arranged closer to the video camera 2 than the persons and / or objects constituting the foreground pixels.

[0020] The first image may refer to any image of a scene captured by the video camera 2, and similarly, the first point in time may refer to any point in time. Similarly, at least the first pixel may refer to any one or more pixels of the image located at any location within the first image. Further, concluding that at least the first pixel is included in the background of the first image may be determined and / or judged by any executable method, for example, detecting pixel values or pixel intensities that are invariant within a predetermined range, such as a known one, for example, IR imaging of the scene, or general image depth detection using a depth camera, etc. That is, the pixels of the image can be regarded as either foreground pixels or background pixels. According to one example, foreground pixels refer to pixels that are dynamic and / or changing, and / or have values, while background pixels may refer to pixels that are static and / or invariant within a predetermined range, and / or have values. The term "identifying" within the image may refer to "detecting, filtering out, and / or determining" within the image throughout the present disclosure, while "within the [···] image" may refer throughout to "within the captured [···] image" and / or "within the [···] image captured by the video camera" throughout. The term "at least the first pixel" may, on the one hand, refer to "one or more pixels" and / or "one or more adjacent and / or dispersed pixels", while "pixels included in the background" may refer to "pixels regarded as and / or determined to be included in the background", "pixels determined to be included in the determined background", "pixels belonging to the background", and / or "pixels included in the background region". According to one example, the term "pixels included in the background of the [···] image" may refer throughout the present disclosure to "pixels included in the static and / or non-changing region of the [···] image" and / or "pixels determined to be static and / or non-changing and / or not changing to a predetermined extent and / or regarded as such".

[0021] The background blurring system 1 may be adapted and / or configured to determine a blurred pixel value for at least a first pixel, for example, by means of a first blurring unit 102. Therefore, a blurring value may be determined for at least a first identified background pixel, and subsequently, pixels determined to be blurred, out of focus, and / or unclear may be provided.

[0022] The determination of the blurred pixel value of one or more identified background pixels can be achieved in any, for example, known and executable method, for example, using the support from one or more generally known image filter kernels, for example, a blurring kernel. Such exemplary image filter kernels may be of any feasible dimension and / or size, for example, square kernels such as 3×3 kernels, 11×11 kernels, 21×21 kernels, etc., and / or non-square kernels such as 21×11 kernels, etc., where 21 may be arranged horizontally. The range by which at least the first pixel is blurred can be arbitrarily selected, for example, considered suitable for the current implementation. Optionally, for bitrate efficiency, the kernel size of the image filter kernel used to determine the blurred pixel value of at least the first pixel may be matched to the macroblock size of the I-frame of the video compression stream. The term "determine" for the blurred pixel value may, throughout this disclosure, also refer to "apply" the blurred pixel value and / or "calculate" the blurred pixel value. Further, according to one example, it may refer to "determine based on an image filter kernel" for the blurred pixel value. "Blurred pixel value", on the one hand, refers to "out-of-focus and / or unclear pixel value", and "for at least a first pixel" may refer to "applicable for at least a first pixel". According to one example, "blurred pixel value" may additionally or alternatively refer to "pixel value with color removed". Further, according to one example, "blur" may, throughout this disclosure, refer to "blurred and / or modified", or simply "modified", and thus, the "modified" pixel value may be modified in any executable method.

[0023] The background blurring system 1 may be adapted and / or configured, for example, by means of the first providing unit 103, to provide the blurred pixel values in the video compression stream and subsequently at a first point in time. Therefore, the determined blurred pixel values of at least the first pixels related to the first captured image are utilized for the video compression stream and, as shown in FIG. 1a, for at least the first stream pixels 111 in the corresponding first compressed image 11 of the stream. Thus, by using the blurred pixel values for one or more identified background pixels in the video compression stream instead of non-blurred pixel values, the bitrate can be reduced and / or privacy issues can be satisfied. Further, among those background pixels, there may be times when they are blurred as opposed to non-blurred foreground pixels, and an object (e.g., a moving person and / or object) including the foreground pixels may be in focus and / or highlighted compared to a non-moving person and / or object including the background pixels. Taking FIG. 1a as an example, all pixels of the first image, including at least the first pixels, are considered to belong to the background. Thus, in the first compressed image 11, all pixels are blurred.

[0024] The blurred pixel values may be provided arbitrarily in the video compression stream, - for example, known -, executable methods, for example, display, - for example, background blur system 1, and / or one or more displays 3 that communicate directly or indirectly with the video camera 2 -, and / or one or more display devices 3 that communicate directly or indirectly with the video camera 2, are displayed, and / or, for later use, for example, recorded, and / or digitally stored. The phrase "providing" [···] pixel values in the video compression stream means, throughout this disclosure, "inserting, setting, utilizing, and / or using" [···] pixel values in the video compression stream, and / or "storing, saving, and / or recording" for use in the video compression stream, and further, according to one example, "providing" [···] pixel values continuously and / or intermittently in the video compression stream, and / or "displaying" and / or "utilizing" [···] pixel values in one or more subsequent frames of the video compression stream. The phrase "providing subsequent to the time point of the [···]" may mean, throughout, "providing subsequent to and / or essentially at the time point of the [···]".

[0025] The background blur system 1 may be adapted and / or configured, for example, by means of the second identification unit 104, to identify that, in the second image of the scene, at a subsequent second time point, at least the first pixel has changed to be included in the foreground of the second image. Thus, it is determined that, in the second image of the scene occurring at any second time point, one or more pixels previously identified as belonging to the background have changed to belong to the foreground region of the captured second image. According to one example, this is caused, for example, by one or more persons and / or objects entering the scene.

[0026] The second image may refer to any image of a scene captured by the video camera 2 after the previously captured first image, and similarly, the second time point may refer to any time point occurring after the first time point, such as from a few milliseconds to several minutes, or even several hours later. It should be noted that the second image is not necessarily the image immediately following the first image, and conversely, there may be any number of consecutive images captured between the so-called first and second images. Further, the conclusion that at least the first pixel is included in and / or was included in the foreground of the second image is determined and may be determined by any, for example, known, method of detecting changes and corresponding changes in pixel values and / or pixel intensities, and / or general image depth detection as previously shown. The phrase "a pixel is changed to be included in the foreground" may refer to "a pixel is changed to be included in the foreground", and / or "a pixel is determined to be included in the foreground", and / or "a pixel has changed to be included in the foreground", while the phrase "a pixel is changed to be included in the foreground" may refer to "a pixel is changed to be included in the determined foreground", "a pixel is changed to belong to the foreground", and / or "a pixel is changed to be included in the foreground region". According to one example, the phrase "a pixel is changed to be included in the foreground of the second image" may refer to "a pixel is changed to be included in the dynamic, changing, and / or significantly changing region of the second image", "a pixel is determined and / or considered to have a - dynamically, changing, and / or significantly changing - value", and / or "a pixel currently has or essentially currently has a - dynamically, changing, and / or significantly changing - value". Further, "the second image" may refer to "the next second image", and / or "the next second image of the first image".

[0027] The background blurring system 1 may be adapted and / or configured, for example, by means of the second providing unit 105, subsequently at a second time point, -continuously and / or intermittently- to provide the identified non-blurred pixel values for at least the first pixel to a video compression stream. Therefore, the identified pixel values of at least the first pixel associated with the second captured image, -and the further identified changing pixel values in any number of subsequent captured images of the scene- are utilized for the corresponding at least first stream pixel 111 of the video compression stream, which is displayed in the corresponding second compressed image 12 of the stream, as shown in FIG. 1b. Thus, by utilizing the non-blurred pixel values, for one or more identified foreground pixels within the video compression stream, the foreground pixels, -i.e., for example, one or more moving persons and / or objects- may be in focus compared to the blurred background pixels. Illustrating FIG. 1b, at least the first pixel of the second image is considered to belong to the foreground and, thus, in the second compressed image 12, the pixel, -here exemplarily belonging to a moving person- is non-blurred, i.e., sharp. In the illustrated example, it should be noted that at least the first pixel may change continuously and / or intermittently as the person continues to move, whereby the non-blurred pixel values for at least the first pixel may change continuously and / or intermittently along with the continuous captured images of the scene.

[0028] The non-blurred pixel value of at least the first pixel may be identified and / or derived in any, for example, known, executable way, for example using the support from generally known image processing. The phrase "providing the identified non-blurred pixel value for [···]" may refer to "providing the captured and / or derived non-blurred pixel value for [···]", "providing the updated non-blurred pixel value for [···]", and / or "providing the updated identified non-blurred pixel value for [···]". Further, according to one example, the phrase "continuously and / or intermittently providing the identified non-blurred pixel value for the at least first pixel to [···]" may refer to "providing the identified non-blurred pixel value for the at least first pixel to [···] as long as the at least first pixel is included in the foreground of the continuously captured images". The phrase "non-blurred pixel value" may, on the one hand, also refer to "actually detected, in-focus, and / or distinct pixel value".

[0029] The background blurring system 1 is adapted and / or configured, for example by means of the third identification unit 106, to vary in a third image of the scene, subsequently at a third point in time, such that at least the first pixel is included in the background of the third image, and to identify that the at least first pixel value remains unchanged from an intermediate point in time to the third point in time. Thus, in the third image of the scene occurring at any third point in time, after being identified as belonging to the background, one or more pixels that are identified as belonging to the foreground are established to be considered as having changed to belong to the background region again after remaining unchanged after the intermediate point in time. According to one example, this may be due to, for example, one or more persons and / or objects leaving the scene and / or being at least partially stationary.

[0030] The third image may refer to any image of the scene captured by the video camera 2 in a subsequent previously captured second image, and similarly, the third time point may refer to any time point that occurs subsequent to the second time point, such as from a few milliseconds to a few minutes, or even several hours later. The third image does not necessarily have to be an image immediately following the second image. Conversely, in a scenario where at least the first pixel remains in the foreground for a certain period of time, for example, regardless of the presence or absence of a change in pixel values, such as arising from the illustrated moving person, any number of consecutive images between the so-called second and third images may have been captured. It should be noted that the conclusion that at least the first pixel at the third time point is configured in the background of the third image may be determined, for example, by detecting, deriving, and / or learning in an - arbitrary - known, - feasible way, that - at least the pixel value of the first pixel is invariant, - and / or essentially invariant, - up to a pre-determinable range immediately preceding the third time point. Here, at least the first pixel is said to remain invariant after an exemplary intermediate time point, which may occur at any time after the second time point, for example, from a few milliseconds to a few minutes, or even several hours later.

[0031] The phrase "the pixel has changed to be included in the background" may refer to "the pixel has changed to be included in the background again", "the pixel has changed to be included in the background", and / or "the pixel has been determined and / or changed to be identified as being included in the background", and further "the pixel has changed to be included in the determined background", "the pixel has changed to belong to the background", and / or "the pixel has changed to be included in the background area". According to an example, "the pixel has changed to be included in the background of the third image" means "the pixel has changed to be included in the static and / or unchanging area of the third image", and / or "the pixel - that has changed to be static and / or unchanging - that has a value - and / or has, and / or within a pre-determinable range, has changed to be static and / or unchanging - that has been determined and / or identified". Further, the term "third image" may refer to "the next third image", and / or "the third image following the second image". Further, the expression "at least the value of the first pixel" may refer to "at least the pixel value of the first pixel", while "remains unchanged" may refer to "essentially remains unchanged", "reflects being non-moving and / or stationary", and / or "remains unchanged within a pre-determinable range". The expression "remains unchanged from the intermediate time point to the third time point" may, on the one hand, refer to "remains unchanged from the intermediate time point to at least the third time point", and / or "remains unchanged from the intermediate time point to the subsequent second time point, the third time point".

[0032] The background blurring system 1 may be adapted and / or configured to provide blurred pixel values within a video compression stream - for example, by means of a third providing unit 107 - subsequently at a third point in time, when the third point in time is within a pre-determined background merge time limit from an intermediate point in time. For this purpose, following said third point in time, from a time range from the intermediate point in time, - for example, when at least a first pixel last changed its pixel value -, the third point in time, - for example, when at least a first pixel changed from a foreground pixel back to a background pixel again -, if the time range up to then is within a set time, the previously determined blurred pixel value for at least the first pixel will, - as shown in Figure 1c -, be reused in the video compression stream for the corresponding at least first stream pixel 111 displayed in the corresponding third compressed image 13 of said stream. That is, when at least a first pixel changes so as to become a background pixel again within a pre-determined period from an invariant foreground pixel, - for example, exemplified by one or more persons and / or objects passing through a scene within said background merge time limit, for example 5 seconds, and / or set not to remain stationary beyond a merge time zone of background pixels -, the already determined blurred pixel value is used again in the video compression stream. Thus, by reusing the already determined blurred pixel value, - considered still valid when the background merge time limit criterion is met -, it is considered unnecessary to determine an updated blurred pixel value for at least the first pixel, and thus reduced computational cost of the video compression stream and / or efficient bitrate compression is provided. Furthermore, since the amount of computation required is reduced, a relatively large image filter kernel, - for example, represented by a blurring kernel -, can be utilized. Illustrating Figure 1c, at least a first pixel of the third image is considered to belong to the background again, and this change occurs, - according to this example -, within the set background merge time limit, and in the third compressed image 13, said pixel is blurred using the previously determined blurred value.

[0033] The background merge time limit may refer to any executable time range, for example, a fixed time range from less than 1 second to tens of seconds, or even up to several minutes or hours. According to one example, the background merge time limit is 2 seconds or more, according to another example it is 5 seconds or more, and according to yet another example it is 10 seconds or more. Similarly, according to one example, the background merge time limit is 60 seconds or less, according to another example it is 30 seconds or less, and according to yet another example it is 10 seconds or less. The background merge time limit may further be variable and / or adjustable, and may be set, for example, to be considered suitable according to the implementation and / or situation. The expression "when the third point in time is within a pre-determinable background merge time limit from the intermediate point in time" means "if the third point in time is within a pre-determinable background merge time limit from the intermediate point in time, what should be provided / or is provided", while "within the pre-determinable background merge time limit" may refer to "within the pre-determinable background merge time limit", and / or simply "within the background merge time limit". Further, the term "background merge time limit" may also refer to the "first time limit, the'reuse threshold time limit'. According to one example, the statement "subsequently at the third point in time, provide the blurred pixel value within the video compression stream" may also refer to "subsequently at the third point in time, reuse the blurred pixel value within one or more subsequent frames of the video compression stream".

[0034] Optionally, the background blur system 1 may be applied and / or configured to determine an updated blurred pixel value for at least a first pixel - for example, by means of an optionally second blur unit 108 - when a third time point is not within the background merge time limit. Thus, from an intermediate time point - that is, from a time point when at least the first pixel value remains unchanged - to a time range up to a third time point - that is, until at least the first pixel, which is a foreground pixel, changes back to a background pixel again - if the time range is not within the set time, an updated blur value for at least the first pixel may be determined. That is, when the criterion of the background merge time limit is not met, determining an updated blurred pixel value for at least the first pixel is considered appropriate in that the previously determined blurred pixel value is considered no longer relevant, as at least the first pixel is considered to reflect a different, new, and / or updated background compared to the previously determined blurred pixel value. The phrase "when a third time point is not within the background merge time limit" may mean that the background merge time limit should be provided and / or is provided that the third time point is not within the background merge time limit, while "updated blurred pixel value" may refer to "new blurred pixel value".

[0035] Optionally, the background blur system 1 may be adapted and / or configured to provide updated blurred pixel values within the video compression stream at a subsequent third point in time, for example, by means of an optional fourth providing unit 109. Therefore, subsequent to said third point in time, from an intermediate point in time, i.e., from when at least the first pixel value remains unchanged, to the third point in time, i.e., from when at least the first pixel changes from a foreground pixel and then back to a background pixel again, when this is not within the time set as described above, the updated, determined blurred pixel value for at least the first pixel is used in the video compression stream, for example, as shown in FIG. 1e, for the corresponding at least first stream pixel 111 displayed in the corresponding alternative third compressed image 13' of the stream. That is, when the time range from when at least the first pixel becomes an unchanged foreground pixel and then back to a background pixel again is within the set time, for example, as illustrated in FIG. 1d, i.e., from when a person enters the scene and then is at least partially stationary for a period longer than the set background merge time limit, then the determined and updated blurred pixel value may be used in the video compression stream instead of the old, and / or considered old, previously determined blurred pixel value. In the example of FIG. 1e, at least the first pixel is considered to belong to the background again, and this change is, as described above, illustrated, for example, by a person entering the scene and then being at least partially stationary for a period longer than the set background merge time limit, so that in the alternative third compressed image 13', the pixel is blurred again with the determined blurred pixel value.

[0036] Furthermore, optionally, the background blur system 1 may further be applied and / or configured to additionally determine updated blur pixel values for at least additional pixels in the vicinity of the first pixel - for example, by means of an optional second blur unit 108. The background blur system 1 may further be applied and / or configured to additionally provide updated blur pixel values for additional pixels - for example, by means of an optional fourth providing unit 109. Therefore, following the aforementioned third time point, in addition to determining and then providing updated blur pixel values for at least the first pixel in the video compression stream, updated pixel values are also determined and then provided for a pre-determinable portion of the pixels in the vicinity of and / or associated with at least the first pixel.

[0037] Additional pixels in the vicinity of at least the first pixel for which updated blur pixel values are additionally determined may be represented by any realizable dimension portion of the pixels in the vicinity of at least the first pixel, such as the pixels in the vicinity of at least the first pixel, and / or in terms of the number of pixels. The additional pixels are represented by pixels forming a rectangular or square frame around at least the first pixel - for example, 10×10 pixels. At least the first pixel does not necessarily have to be at the center of its rectangular or square frame. According to a further example, the additional pixels may be represented by a part of the object to which at least the first pixel belongs and / or pixels considered to be associated. The phrase "for the additional pixels in the vicinity of the at least first pixel" may refer to "for the additional pixels surrounding and / or in the vicinity of the at least first pixel", and / or "for a part of the additional pixels in the vicinity of the at least first pixel".

[0038] Optionally, the background blur system 1 may be adapted and / or configured to determine, from the current image of the scene, an updated blurred pixel value for at least a first pixel when it is determined that the blurred pixel value at a pre-determined additional time point - for example, by means of an optional additional blur unit 110 - has not been updated over a pre-determined additional time period. The background blur system 1 may further be adapted and / or configured to provide, at a subsequent additional time point - for example, by means of an optional additional provision unit 111 - the updated blurred pixel value into the video compression stream. Thus, following said additional time point, if it is determined at any time that the blurred pixel value has not been updated within a set time limit, then subsequently, the updated blurred pixel value is determined for at least a first pixel from the current image and is then used in the video compression stream. Accordingly, even if at least a first pixel remains a background pixel within a set additional time limit, an updated blurred pixel value may nevertheless be determined and provided into the video compression stream. Thus, for example, situations involving the provision of previously determined blurred pixel values - which may potentially be old and / or incorrect - may be corrected. Similarly, for example, situations which do not necessarily cause at least a first pixel to change from background to foreground - including slowly changing light conditions such as between dusk or dawn - may additionally or alternatively be corrected.

[0039] The additional time limit may be any executable time range, for example, ranging from less than 1 second to dozens of seconds, or even up to several minutes or hours. According to one example, the additional time limit may be represented by a time range equal to or essentially equal to the time range of the aforementioned background merging time limit. The additional time limit may further be variable and / or adjustable, for example, appropriately set according to the nearby implementation and / or situation. Similarly, the additional time point may be any executable time point. According to one example, the additional time point may be represented by a time point that occurs at a time limit equal to or essentially equal to the background merging time limit after the aforementioned additional time point. Note that the additional time point and / or its magnitude may vary by pixel. The phrase "the blurred pixel value at a pre-determinable additional time point" may also refer to "if, among the pre-determinable additional points, the blurred pixel value is to be provided and / or has been provided". The phrase "pre-determinable additional time point" on the one hand refers to "pre-determined additional time point" and / or simply "pre-determinable additional time point", while "pre-determinable additional time limit" may simply refer to "pre-determinable additional time limit" and / or "additional time limit". Further, the phrase "from the current image of the scene" may also refer to "essentially currently and / or from the next image of the scene".

[0040] Optionally, the background blur system 1 may be adapted and / or configured to determine updated blurred pixel values for at least a first pixel and for a further portion of pixels of a predetermined portion of pixels that are identified as being composed within the background of the current image of the scene when a predetermined subsequent frame of the video compression stream is an I-frame, at a time related to the predetermined I-frame, for example, by means of an optional I-frame blur unit 112. The background blur system 1 may further be adapted and / or configured to provide updated blurred pixel values into the video compression stream at a time related to a subsequent I-frame, for example, by means of an optional I-frame providing unit 113. Thus, following the time related to the I-frame, if a predetermined subsequent frame of the video compression stream is an I-frame, the updated blurred pixel values are determined from the current image for at least a first pixel and for a portion of pixels further selected from the background pixels and may then be used within the video compression stream. Accordingly, for one or more selected I-frames, a range of selected background pixels, including at least a first pixel, may be updated. Thus, the updated blurred pixel values are determined and supplied for the selection of pixels without adversely affecting the computational cost and / or the bitrate of the video compression stream, where the pixel selection may vary for each selected I-frame.

[0041] The time points related to the I-frame that can be pre-determined may be represented by any achievable time point, and similarly, the subsequent I-frame that can be pre-determined may refer to any achievable I-frame in the video compression stream. For example, all the nth I-frames may be selected - that is, pre-determinable - I-frames, where n may be selected according to the implementation nearby and / or the situation. The phrase "when the subsequent frame that can be pre-determined in the video compression stream [···] is an I-frame" refers to "in the video compression stream [···], the subsequent frame that can be pre-determined should be supplied with and / or is supplied with an I-frame", while "subsequent frame" may refer to "this frame" and / or "the next frame". "Subsequent frame that can be pre-determined", on the one hand, refers to "subsequent frame that has been pre-determined", and / or "selected subsequent frame", and "time point related to the I-frame that can be pre-determined", on the other hand, may refer to "time point related to the I-frame that has been pre-determined", and / or simply "time point related to the I-frame". Furthermore, the phrase "further pixels of the pre-determinable part" refers to "additional pixels of the pre-determinable part", "further pixels of the pre-determined part", and / or simply "further pixels of the part", while the phrase "part of the pixels identified as being included in the background" may refer to "range, set, and / or selection from the pixels identified as being included in the background". Furthermore, the expression "of the current image of the scene" may also refer to "essentially current, and / or, of the subsequent image of the scene".

[0042] As further shown in FIG. 2, the background blur system 1 may include a first identification unit 101, a first blur unit 102, a first providing unit 103, a second identification unit 104, a second providing unit 105, a third identification unit 106, a third providing unit 107, an optional second blur unit 108, an optional fourth providing unit 109, an optional additional blur unit 110, an optional additional providing unit 111, an optional I-frame blur unit 112, and an optional I-frame providing unit 113, all of which have been described in more detail above. Further, embodiments of the present specification for providing a video compression stream from a video camera 2 adapted to capture a scene are implemented with one or more processors, such as a simple graphics processing unit, a GPU, and / or a central control unit, a CPU, and computer program code for performing the functions and operations in the present embodiment, which is a processor 114. The program code may be provided as a computer program product, for example, in the form of a data carrier carrying the computer program code for executing the embodiments of the present application when mounted on the background blur system 1. One such carrier may be in the form of a CD ROM disk, an optical disk, a solid state disk, a flash memory, and / or a hard drive, but other data carriers are feasible. The computer program code may further be provided as pure computer program code on a server and downloaded to the background blur system 1. The background blur system 1 may further include a memory 115 including one or more memory units. The memory 115 is used, for example, to store information and further store data, configurations, schedulings, and applications, and may be configured to execute the method of the present application when executed in the background blur system 1. For example, the computer program code may be implemented in the firmware stored in the flash memory 115 of the embedded processor 114.Furthermore, the units 101-113, optional processor 114, and / or optional memory 115 may be at least partially included in, associated with, and / or connected to the video camera 2 and / or a monitoring system including, for example, the video camera 2. Those skilled in the art will also understand that the units 101-113 described above may refer to a combination of analog and digital circuits and / or one or more processors configured with software and / or firmware stored in a memory such as the memory 115, and when executed by one or more processors such as the processor 114, will be executed as described herein. One or more of these processors, as well as other digital hardware, may be included in one application specific integrated circuit (ASIC) or multiple processors, and the various digital hardware may be distributed among multiple independent components, whether individually packaged or implemented in a system on chip (SoC).

[0043] FIG. 3 is a flowchart illustrating an exemplary method executed by the background blur system 1 according to an embodiment of the present disclosure. The method is for providing a video compression stream from a video camera 2 configured to capture a scene. An exemplary method that can be continuously repeated includes one or more of the following operations discussed with support from FIGS. 1-2. Further, the operations may be performed in any suitable order and / or one or more operations may be performed simultaneously and / or in an alternative order. For example, optional operations 1010 and 1011 and / or operations 1012 and 1013 may be performed before operations 1008 and 1009.

[0044] Operation 1001 In operation 1001, the background blur system 1 identifies at least a first image included in the background of the first image at a first point in time within the first image of the scene - for example, with support from the first identification unit 101.

[0045] Operation 1002 In operation 1002, the background blurring system 1 determines, for at least the first pixel, a blurred pixel value, for example, with the support from the first blurring unit 102.

[0046] Operation 1003 In operation 1003, the background blurring system 1 provides the blurred pixel value into the video compression stream at a subsequent first time point, for example, with the support from the first providing unit 103.

[0047] Operation 1004 In operation 1004, the background blurring system 1 identifies, in the second image of the scene at a subsequent second time point, for example, with the support from the second identifying unit 104, that at least the first pixel has changed such that it is included in the foreground of the second image of the scene.

[0048] In operation 1005, the background blurring system 1 provides, into the video compression stream at a subsequent second time point, continuously and / or intermittently, for at least the first pixel, the non-blurred pixel value identified, for example, with the support from the second providing unit 105.

[0049] Operation 1006 In operation 1006, the background blurring system 1 identifies, in the third image of the scene at a subsequent third time point, for example, with the support from the third identifying unit 106, that at least the first pixel has changed such that it is included in the background of the third image of the scene and that at least the first pixel value remains unchanged from the intermediate time point to the third time point.

[0050] Operation 1007 In operation 1007, when the third time point is within a pre-determinable background merge time limit from the intermediate time point, the background blurring system 1 provides the blurred pixel value into the video compression stream at a subsequent third time point, for example, with the support from the third providing unit 107.

[0051] Operation 1008 In an optional Operation 1008, when the third time point is not within the background merge time limit, the background blur system 1 determines an updated blurred pixel value for at least the first pixel - for example, with the support from an optional second blur unit 108.

[0052] Optionally, the determination in Operation 1008 includes additionally determining an updated blurred pixel value for additional pixels in the vicinity of at least the first pixel.

[0053] Operation 1009 In an optional Operation 1009, after an optional Operation 1008, when the third time point is not within the background merge time limit, the background blur system 1 subsequently provides an updated blurred pixel value into the video compression stream at the third time point - for example, with the support from an optional fourth providing unit 109.

[0054] Optionally, the provision in Operation 1009 includes additionally providing an updated blurred pixel value for additional pixels.

[0055] Operation 1010 In an optional Operation 1010, when it is determined that the blurred pixel values at pre - determinable additional time points have not been updated over pre - determinable additional time points, the background blur system 1 determines an updated blurred pixel value for at least the first pixel from the current image of the scene - for example, with the support from an optional shooting blur unit 110.

[0056] Operation 1011 In an optional Operation 1011, after an optional Operation 1010, when it is determined that the blurred pixel values at pre - determinable additional time points have not been updated over pre - determinable additional time points, the background blur system 1 provides an updated blurred pixel value into the video compression stream at a subsequent additional time point - for example, with the support from an optional additional providing unit 111.

[0057] Operation 1012 In an optional operation 1012, when a predictable subsequent frame of the video compression stream is an I-frame at a predictable I-frame related time point, the background blurring system 1 determines, for example, with the support from an optional I-frame blurring unit 112, at least a first pixel among the pixels identified as being included in the background of the current image of the scene, and an updated blurred pixel value for a predictable portion.

[0058] Operation 1013 In an optional operation 1013, after the optional operation 1012, when a predictable subsequent frame of the video compression stream is an I-frame at a predictable I-frame related time point, the background blurring system 1 subsequently provides, for example, with the support from an optional I-frame providing unit 113, an updated blurred pixel value into the video compression stream at an I-frame related time point.

[0059] Those skilled in the art will understand that the present disclosure is in no way limited to the above preferred embodiments. On the contrary, many modifications and variations are possible within the scope of the appended claims. Further, it should be noted that the drawings are not necessarily to scale and dimensions, and features may be exaggerated at times for clarity. Instead, emphasis is placed on explaining the principles of the embodiments herein. Further, in the claims, the term "comprising" does not exclude other elements or steps, and the indefinite article "a" or "an" does not exclude a plurality.

Claims

1. A method performed by a background blur system for providing a video compression stream from a video camera configured to capture a scene, the method comprising: identifying at least a first pixel included in a background of the first image of the scene at a first point in time in the first image of the scene; determining a blurred pixel value for the at least first pixel; subsequent to the first point in time, providing the blurred pixel value to the video compression stream; identifying at a subsequent second point in time in a second image of the scene that the at least first pixel has changed to be included in a foreground of the second image; subsequent to the second point in time, providing to the video compression stream, continuously and / or intermittently, an unblurred pixel value identified for the at least first pixel; identifying at a subsequent third point in time in a third image of the scene that the at least first pixel has changed to be included in a background of the third image when the at least first pixel has changed from a foreground pixel back to a background pixel; wherein the third point in time is within a predetermined background merge time from an intermediate point in time subsequent to the second point in time, and when at least the first pixel value remains unchanged from the intermediate point in time, providing the blurred pixel value to the video compression stream subsequent to the third point in time comprising a method.

2. when the third point in time is not within the background merge time, determining an updated blurred pixel value for the at least first pixel; and subsequent to the third point in time, providing the updated blurred pixel value to the video compression stream The method according to claim 1, further comprising.

3. Determining the updated blurred pixel value includes additionally determining an updated blurred pixel value for additional pixels near the at least first pixel, and providing the updated blurred pixel value includes additionally providing the updated blurred pixel value for the additional pixels. The method according to claim 2.

4. Determining an updated blurred pixel value for the at least first pixel from the current image of the scene when it is determined that the blurred pixel value at a pre-determinable complement point in time has not been updated during a pre-determinable complement time; and subsequent to the complement point in time, providing the updated blurred pixel value to the video compression stream The method according to claim 1, further comprising

5. At a pre-determinable I-frame related time point, when the pre-determinable next frame of the video compression stream is an I-frame, determining updated blurred pixel values for the at least first pixel and for further pixels of a pre-determinable part of the pixels identified as being included in the background of the current image of the scene; subsequent to the I-frame related time point, providing the updated blurred pixel values to the video compression stream The method according to claim 1, further comprising

6. The method according to claim 1, wherein a kernel size of an image filter kernel used to determine a blurred pixel value of the at least first pixel matches a macroblock size of an I-frame of the video compression stream.

7. A storage medium storing a computer program including computer program code which, when executed by a computer or a processor, causes the computer or the processor to execute the method according to claim 1.

8. A background blurring system for providing a video compression stream from a video camera configured to capture a scene, the background blurring system comprising: a first identification unit for identifying at least a first pixel included in the background of a first image of the scene at a first time point in the first image; a first blurring unit for determining a blurred pixel value for the at least first pixel; a first providing unit for providing the blurred pixel value to the video compression stream subsequent to the first time point; a second identification unit for identifying that the at least first pixel has changed such that it is included in the foreground of a second image of the scene at a subsequent second time point in the second image of the scene; a second providing unit for providing, subsequent to the second time point, continuously and / or intermittently, non-blurred pixel values identified for the at least first pixel to the video compression stream; a third identification unit for identifying that the at least first pixel has changed such that it is included in the background of a third image of the scene when the at least first pixel changes from a foreground pixel back to a background pixel at a subsequent third time point in a third image of the scene; When the third time point is within a predetermined background merge time from an intermediate time point following the second time point and at least the first pixel value remains unchanged from the intermediate time point, a third providing unit that provides the blurred pixel value to the video compression stream following the third time point A background blurring system comprising: **Claim 9** When the third time point is not within the background merge time, A second blurring unit that determines a blurred pixel value updated for at least the first pixel; A fourth providing unit that provides the updated blurred pixel value to the video compression stream following the third time point The background blurring system according to claim 8, further comprising: **Claim 10** The second blurring unit is configured to additionally determine an updated blurred pixel value for additional pixels near at least the first pixel, and the fourth providing unit is configured to additionally provide the updated blurred pixel value for the additional pixels. The background blurring system according to claim 9. **Claim 11** A complementary blurring unit that determines an updated blurred pixel value for at least the first pixel from the current image of the scene when it is determined that the blurred pixel value at a pre-determinable complementary time point has not been updated during a pre-determinable complementary time; A complementary providing unit that provides the updated blurred pixel value to the video compression stream following the complementary time point The background blurring system according to claim 8, further comprising: **Claim 12** At a pre-determinable I-frame related time point, when a pre-determinable next frame of the video compression stream is an I-frame, An I-frame blurring unit that determines an updated blurred pixel value for at least the first pixel and additional pixels of a pre-determinable portion of the pixels identified as being included in the background of the current image of the scene; An I-frame providing unit that provides the updated blurred pixel value to the video compression stream following the I-frame related time point The background blurring system according to claim 8, further comprising: **Claim 13** The background blurring system according to claim 8, wherein a kernel size of an image filter kernel used to determine a blurred pixel value of at least the first pixel matches a macroblock size of an I-frame of the video compression stream. **Claim 14** A video camera comprising the background blur system according to claim 8.

15. A background blur system for providing a video compression stream from a video camera configured to capture a scene, the background blur system comprising: At a first point in time in a first image of the scene, identifying at least a first pixel included in the background of the first image; Determining a blurred pixel value for the at least first pixel; Subsequent to the first point in time, providing the blurred pixel value to the video compression stream; At a subsequent second point in time in a second image of the scene, identifying that the at least first pixel has changed to be included in the foreground of the second image; Subsequent to the second point in time, providing an unblurred pixel value identified for the at least first pixel to the video compression stream; At a subsequent third point in time in a third image of the scene, identifying that the at least first pixel has changed to be included in the background of the third image when the at least first pixel has changed back from a foreground pixel to a background pixel again; When the third point in time is within a predetermined background merge time from an intermediate point in time following the second point in time and at least the first pixel value remains unchanged from the intermediate point in time, providing the blurred pixel value to the video compression stream subsequent to the third point in time A background blur system composed of a circuit.

Images