Method, device, system, and storage medium for displaying video stream

Abstract

The present disclosure relates to displaying a video stream. A computer-implemented method is provided for displaying a video stream of a captured scene on a display. The method comprises detecting a new event in the scene, determining a new event region in the scene within which the new event is detected, checking whether a prior event has been detected within the new event region during a predetermined time period preceding the new event, adding a visual indicator to an image frame corresponding to a time of detection of the new event and to a plurality of subsequent image frames of the video stream in the absence of a prior event being detected during the predetermined time period, wherein the visual indicator coincides with the new event region, and wherein the step of adding the visual indicator comprises gradually changing an appearance of the visual indicator throughout the plurality of subsequent image frames, and displaying the video stream with the added visual indicator on a display of a client device. Related client devices and non-transitory computer-readable storage media are also provided.

Description

Technical Field

[0001] The present invention relates to a method and apparatus for displaying a video stream with added indicators that indicate events in the video stream to an operator or viewer of the video stream. Background Art

[0002] Surveillance cameras can both prevent crime by deterring potential criminals from committing crimes and provide investigators with important forensic information about crimes that have already been committed. Surveillance cameras can also be used for security purposes to monitor potentially dangerous environments (such as railroad crossings). Often, it is preferable to form a large system of multiple independent surveillance cameras to cover a larger area, rather than using just a single camera. While such a surveillance camera system can provide a large number of video streams covering everything intended to be monitored, the large number of video streams can require significant human resources and attention to actually detect important details within the video stream when viewing it on a display. Adding bounding boxes and markers to the video stream to indicate, for example, objects or motion, can help human operators detect movement that might otherwise have been missed. However, creating such indicators for all movement can be disadvantageous for encoding, transmitting, and storing the video stream. Furthermore, if multiple movements exist in the scene covered by a surveillance camera, the operator will inevitably become distracted and fatigued by the constant appearance of new indicators on the display. Therefore, such solutions do not effectively address the issue of requiring human resources and attention. Therefore, improvements are needed in the art. Summary of the Invention

[0003] One object of the present invention is to conserve computer processing and storage resources.Another object is to increase the likelihood that a human operator will detect important events in a video stream that are not readily apparent or that are perhaps obscured by other events.

[0004] According to a first aspect, a computer-implemented method is provided for displaying a video stream of a scene captured by a video capture device on a display of a client device. The method comprises: detecting a new event in the scene; determining a new event region in the scene within which the new event is detected; and checking whether a prior event has been detected within the new event region during a predetermined time period before the new event. When no prior event is detected during the predetermined time period, the method further comprises adding a visual indicator to an image frame corresponding to the detection time of the new event and to a plurality of subsequent image frames of the video stream, wherein the visual indicator is consistent with the new event region. The step of adding the visual indicator comprises gradually changing the appearance of the visual indicator throughout a plurality of subsequent image frames. The method further comprises displaying the video stream with the added visual indicator on the display of the client device.

[0005] The term client device refers to any type of device that has the ability to display or functionality associated with displaying a video stream. The term event can refer to both an event that progresses in time or an instantaneous event. An event can be characterized, for example, as movement in a scene and therefore also as movement between image frames in a video stream depicting the scene. An event can additionally or alternatively be understood as an object present in a scene captured by a video capture device. An object can be an object of interest. An object of interest can be an object from a particular class of objects, such as a human, a face, a car, etc. During an event, an object typically exists during multiple image frames of the video stream of the scene. The term new event can refer to an event that is new for a particular image frame compared to the previous image frame. A new event can then be considered to last for multiple image frames. The term event area can refer to an area within a scene associated with a particular event.

[0006] The term "corresponding in time to..." refers to the temporal correspondence of, for example, an event, indicator, or image frame. Specifically, by an image frame corresponding in time to the detection of a new event, it is meant an image frame acquired at the same time as the detection of the new event. The same time is understood to include a time margin from when the image frames were acquired in a separate, non-continuous manner. A temporally corresponding image frame can therefore be understood to be an image frame acquired at a time closest to (just before or just after) the time point of the detection of the new event.

[0007] The inventors have recognized that introducing a requirement for event indication in video streams may be advantageous for a number of different reasons. For one, events can be filtered so that only those suspected of being important or relevant are indicated or highlighted in the video stream. This can conserve computer processing resources and thereby reduce the requirements on the hardware required to encode and decode the video stream. Similarly, it can conserve storage resources and reduce the requirements on computer storage media and data transmission channels.

[0008] Additionally, the requirement for event indications may reduce fatigue for a human operator viewing a video stream and increase the likelihood that the operator will detect important or relevant events in the video stream.

[0009] An advantage associated with the requirement that no prior events be detected during the predetermined time period is that events in areas of the video stream that see low event activity (e.g., low levels of motion between image frames) are likely to be missed without the operator viewing the video stream noticing them. This may be due to the operator being more focused on areas of the displayed video that are experiencing more events.

[0010] Even when a human operator is initially focused on other areas, the requirement that no events have been detected in a particular area of ​​the video stream during a predetermined period of time to indicate an event feature at least increases the probability that the human operator will direct their attention to the indicated area. This requirement can alternatively be considered as a filter for determining which events to add visual indicators to. Overall, the methods of the present disclosure that describe filtering events to be visually indicated provide benefits because data source requirements can be reduced and operators can be directed more efficiently.

[0011] The feature of gradually changing the appearance of the visual indicator can further increase the probability that the human operator will direct attention to the indicated area. The gradual change in appearance can also provide additional information to the human operator, such as an indication of how much time has passed since the event that triggered the indicator occurred, and thus help the human operator quickly detect the recent event.

[0012] The method may further comprise determining a duration from a last preceding event in the new event region.The number of subsequent image frames and / or the appearance of the visual indicator may be based on the duration.

[0013] In this way, more unexpected events occurring in an area at a longer time that has passed since the last prior event occurring in the new event area can be provided with, for example, more prominent visual indicators (which remain visible for a longer period of time and / or have a more prominent appearance), thereby further increasing the probability that the operator will direct attention to the indicated area.

[0014] The method may further include detecting an ongoing discrete event in the scene concurrently with the new event. Upon detecting the discrete event, the method may further include determining a discrete event region in the scene within which the discrete event was detected. The method may further include determining a distance between the new event region and the discrete event region. Upon determining that the distance between the new event region and the discrete event region is greater than a predetermined threshold, the step of adding a visual indicator may be performed. The discrete event in the scene may be detected by the same video capture device that captured the new event or by another video capture device.

[0015] An advantageous effect of taking into account the distance of one or more other events is that an operator viewing the video stream can easily detect a new event that occurs very close to a discrete event that is already in progress, without requiring a large and striking visual indicator, further conserving data processing resources. This is likely because the operator's attention has presumably already been directed to the discrete event area and the operator will therefore be more likely to detect the proximate new event.

[0016] The number of subsequent image frames and / or the appearance of the visual indicator may be based on the distance between the new event region and the discrete event region.

[0017] As such, new events that are further away from the discrete event in the video stream can be provided with, for example, more prominent visual indicators (which persist longer or have a more prominent appearance), thereby further increasing the probability that the operator will direct their attention to the indicated area. Alternatively, in certain circumstances, new events that are closer to the discrete event can be provided with a more prominent appearance because otherwise these events might be obscured by the discrete event that captures all or too much of the operator's attention.

[0018] The method may further comprise determining a motion level of the new event. Upon determining (ie assuming) that the motion level is greater than a predetermined threshold, the step of adding a visual indicator may be performed.

[0019] By introducing a motion level requirement for new events, new motion events can be filtered based on their motion to add visual indicators. This can be advantageous in applications where the motion level is related to the importance or relevance of the event within the video stream. By filtering based on motion, fewer data sources can be used and operator attention can be drawn more efficiently.

[0020] The method may further include determining a motion level of the new event.The number of subsequent image frames and / or the appearance of the visual indicator may be based on the motion level.

[0021] As such, new events with more motion may be provided with, for example, more prominent visual indicators (which remain visible for a longer period of time and / or have a more prominent appearance), thereby further increasing the probability that the operator will direct attention to the indicated area.

[0022] According to some embodiments, the method may further comprise: determining an object classification for the object, the object being associated with the new event.The step of adding the visual indicator may be performed based on the determined object classification.

[0023] In this paper, the term object refers to any object or structure that can be detected in a video stream. The term object classification refers to a method of grouping similar objects into the same category as long as they share a certain number of common features.

[0024] According to some embodiments, the method may further comprise determining a size of an object associated with the new event.The step of adding a visual indicator may be performed based on the determined size of the object.

[0025] The term size may refer herein to the size of an object in an image frame of a video stream or the predicted real-world size of an object.

[0026] By determining the object associated with the new event, additional options are available for optimizing event filtering and directing operator attention to more important or relevant objects. For example, if the new event is determined to correspond to an object such as a human appearing in the scene, a visual indicator can be added corresponding to the depicted object in the video stream. However, if the new event is determined to correspond to an object such as a bird appearing in the scene, that object can be excluded from receiving the visual indicator in the video stream.

[0027] By determining the size of objects associated with new events, further options are made available for optimizing the filtering of events and directing operator attention to more important or relevant objects.

[0028] The visual indicator may be configured to follow the movement of the new event across subsequent image frames of the video stream.

[0029] Such embodiments may provide an improved ability for an operator viewing a video stream to detect and track new events that move between subsequent image frames of the video stream.

[0030] The step of gradually changing the appearance of the visual indicator may include one or more of: gradually expanding the visual indicator throughout multiple subsequent image frames, gradually fading the visual indicator, gradually changing the form of the visual indicator, gradually rotating the visual indicator, and gradually changing the color of the visual indicator.

[0031] Different types of gradually changing visual indicators can serve different purposes and provide additional information about the indicated event to an operator viewing the video stream without necessarily using a text message. The different types of appearance changes can depend on the predicted importance or relevance of the indicated event. The visual indicator can thus be adapted to at least influence the operator to direct an appropriate level of attention to each indicated event, based on, for example, the importance or relevance of the indicated event.

[0032] The method may further include determining a viewing point of at least one eye of an operator viewing the display. Adding a visual indicator may be performed when the viewing point is determined to be outside the display or when the distance between the viewing point and the new event area is determined to be greater than a predetermined threshold.

[0033] The term viewing point refers to the point in space within or outside the display that the operator's eyes are detected or predicted to be directed towards or focused on. The viewing point may alternatively be understood as the point of attention of an operator viewing a video stream on the display.

[0034] In this way, new events can be indicated based on where the operator's attention is directed. For example, new events close to the viewing point can be excluded from the indication, while new events far from the viewing point can be indicated by adding a visual indicator. Thus, the filtering of events to be indicated and the guidance of the operator's attention are further optimized.

[0035] The method can be performed on a live stream (e.g., a video stream displayed to an operator) or on a recorded stream that is stored for later review. The method can be performed on stored video data while being recorded or at a later time. The method is therefore not limited to either live streaming / viewing or review of recorded video data. According to a second aspect, there is provided a non-transitory computer-readable storage medium having instructions stored thereon for performing the computer-implemented method according to the first aspect when the instructions are executed on a device having processing capabilities.

[0036] Such an aspect provides similar advantages as the first aspect, and implements the computer-implemented method discussed above.

[0037] According to a third aspect, a client device for displaying a video stream of a scene is provided. The client device includes control circuitry configured to perform an event detection function, the event detection function being configured to detect a new event in the scene. The control circuitry is further configured to perform an event location function, the event location function being configured to determine a new event region in the scene corresponding to a region within which the new event is detected. The control circuitry is further configured to perform an event comparison function, the event comparison function being configured to check whether a prior event has been detected within the new event region during a predetermined time period prior to the new event. The control circuitry is further configured to perform an indicator addition function, the indicator addition function being configured to add a visual indicator to an image frame and to multiple subsequent image frames of the video stream if no prior event has been detected during the predetermined time period. The visual indicator is consistent with the new event region. The indicator addition function is further configured to gradually change the appearance of the visual indicator throughout the multiple subsequent image frames. The control circuitry is further configured to perform a display function, the display function being configured to display the video stream with the added visual indicator on a display of the client device.

[0038] Functions described herein may refer to devices that physically perform the functions they are configured for. Functions may alternatively be implemented as instructions stored on a computer-readable storage medium, which may be executed or performed by a computer or device with processing capabilities.

[0039] Such an aspect relating to a client device for displaying a video stream is characterized by similar objects and advantages as the first aspect.

[0040] According to a fourth aspect, there is provided a system comprising a client device according to the third aspect and a video capture device.

[0041] Such a camera-included system enables advantageous use as a client device for a video surveillance / monitoring system that feeds a captured video stream from a video capture device to the client device which can be displayed to a human operator.

[0042] In any disclosed aspect, the video capture device may be a surveillance camera for wide-angle imaging, a surveillance camera for panoramic imaging, or a surveillance camera for 360-degree imaging, or a portion of multiple video capture devices that jointly perform any of the various imaging mentioned.

[0043] The system according to the fourth aspect can be particularly advantageous in covering wide, large, and / or panoramic scenes. Surveillance cameras may often be required to cover such scenes, sometimes at the expense of detectability of events in the scene. As such, a system that also includes client devices that add visual indicators can help cover large scenes without sacrificing event detectability or at least mitigate the problem of lost detectability.

[0044] Further scope of application of the present invention will become apparent from the detailed description given below. However, it should be understood that the detailed description and specific examples, while indicating preferred embodiments of the present invention, are given as illustrations only, since various changes and modifications within the scope of the present invention will become apparent to those skilled in the art based on the detailed description. It is further noted that the present invention relates to all possible combinations of features, unless expressly stated otherwise.

[0045] Therefore, it should be understood that the present invention is not limited to the specific components of the apparatus described or the behavior of the methods described, as such apparatus and methods may vary. It should also be understood that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] The above and other aspects of the present invention will be described in more detail hereinafter with reference to the accompanying drawings.The drawings should not be considered limiting; rather they should be considered for explanation and understanding purposes.

[0047] As illustrated in the figures, the sizes of layers and regions may be exaggerated for illustrative purposes and, therefore, are provided to illustrate the general structure. Throughout this document, the same reference numerals refer to the same elements. Flowcharts where a step box is delimited by a dashed line indicate that the step is optional for the illustrated embodiment.

[0048] Figure 1A flow chart illustrating a method of displaying a video stream with visual indicators.

[0049] Figures 2a to 2d The diagram shows four different image frames of a video stream and how visual indicators can be added based on events occurring therein.

[0050] Figure 3a to Figure 3b The diagram illustrates how visual indicators can be added based on the distance between new and other ongoing discrete events.

[0051] Figure 4 A flow chart illustrating a method of displaying a video stream with visual indicators based on distances between new and other discrete events.

[0052] Figure 5a to Figure 5b The diagram illustrates how visual indicators may be added depending on the motion level of new events.

[0053] Figure 6 A flow chart illustrating a method of displaying a video stream with visual indicators based on the motion level of new events.

[0054] Figures 7a and 7b The diagram illustrates how visual indicators may be added based on the object classification or size of the object associated with the new event.

[0055] Figure 8 A flow chart illustrating a method of displaying a video stream with visual indicators based on object classification or size of objects associated with new events.

[0056] Figures 9a to 9b The diagram shows two different image frames of a video stream and how a visual indicator may be configured to follow the movement of a new event.

[0057] Figures 10a to 10b Illustrates how a visual indicator may be added based on the distance between the new event and the viewing point.

[0058] Figure 11 A method of displaying a flowchart of a video stream with visual indicators based on the distance between a new event and a viewing point is shown.

[0059] Figure 12 The diagram illustrates a client device for displaying a video stream of a scene in a system with a video capture device that captures the video stream. DETAILED DESCRIPTION

[0060] The present invention will now be described more fully hereinafter with reference to the accompanying drawings, in which presently preferred embodiments of the invention are shown. However, the present invention may be embodied in many different forms and should not be construed as limited to the embodiments set forth herein; rather, these embodiments are provided for thoroughness and completeness, and to fully convey the scope of the invention to those skilled in the art. In addition to the embodiments specifically disclosed herein, any combination of the embodiments is possible, unless expressly stated otherwise.

[0061] Now refer to Figure 1 as well as Figures 2a to 2d To disclose embodiments of the present invention.

[0062] A computer-implemented method is provided for displaying a video stream depicting a scene 100 captured by a video capture device on a display 101 of a client device. Figure 1 Provide a step-by-step demonstration of the method. Please note that the disclosed order of the steps in this disclosure (particularly in the figures) is intended only as a non-limiting example. The steps can be performed in other orders than those disclosed or illustrated.

[0063] like Figure 1 As shown in FIG, the method detects S1002 a new event 112 in the scene 100. The step of detecting S1002 a new event 112 may be based on and performed by an algorithm of computer image analysis, motion detection, object recognition, radar detection (e.g., Doppler radar), etc. Note that the term video stream refers to a depiction of the scene 100. Therefore, in this document, these terms should be considered interchangeable.

[0064] A video stream includes a plurality of image frames 103. Each image frame 103 corresponds to a discrete point in time when that particular image frame 103 was captured. A video stream can be characterized by a specific number of image frames 103 per second (FPS). Preferably, the FPS number is in the range of 1-100, more preferably in the range of 10-60, and most preferably in the range of 20-30. Image frames 103, and therefore the video stream, can represent a panoramic, stitched, or 360-degree image of a scene 100.

[0065] New events 112, and generally other types of events described herein, can be progressive in time (i.e., spanning at least two image frames) or instantaneous (i.e., occurring in only one image frame). Events can involve instantaneous / sudden events or more gradual / flowing events. Events can involve motion between subsequent image frames in a video stream. Such motion can be described as a larger object depicted by a large number of pixels moving in the video stream or as the smallest possible motion demonstrated by a single pixel changing its color value, and anything in between. New events 112 may not necessarily be new in the context of the entire duration of the video stream. New events 112 may be newer in a temporally relative context compared to prior events. More than one new event 112 can be detected S1002 within a single image frame 103.

[0066] An event may also refer to an event that is not visible in the video stream. Such an event may still be motion, but is detected by, for example, a radar acquisition device that also covers scene 100 (i.e., has the same or at least overlapping field of view). The method further includes determining (S1004) a new event region 114 in scene 100 in which new event 112 is detected. New event region 114 may correspond to new event 112 in the video stream in terms of size or location.

[0067] The new event area 114 may, for example, correspond to an area outlined by pixels associated with the new event 112 that have moved between image frames 103 associated with the new event 112. The new event area 114 may correspond to an arbitrary geometric shape and may not necessarily correspond to the new event 112 in terms of size or position. The new event area 114 may include a fill area in addition to the actual spatial extent of the new event 112 detected S1002 in the image frame 103. Throughout this disclosure, spatial extent may refer to an area in an image frame corresponding to a particular event such as the new event 112. Spatial extent may also be used for other types of events disclosed herein. The new event area 114 may therefore include some of the background that typically represents a fixed scene area. More than one new event area 114 may be determined S1004 within a single image frame 103.

[0068] If a new event 112 is determined to have moved between subsequent image frames 103, a new event area 114 may be configured to follow the new event 112. This may be due to an actual new event 112 moving in the scene 100 or due to camera motion such as pan, tilt, pan, zoom, or focus.

[0069] The method further comprises checking S1006 whether a preceding event has been detected within the new event area 114 during a predetermined period of time before the new event 112. A preceding event may for example be understood as an event that may have been the new event 112 at a previous point in time.

[0070] The step of checking S1006 for prior events can include analyzing a list, array, or lookup table to find prior events stored therein. To this end, the method can further include storing the new event 112 as a prior event in the list, array, or lookup table for future reference. The prior event can be stored only temporarily in the list, array, or lookup table to conserve data storage resources.

[0071] Furthermore, a non-limiting alternative to the step of checking S1006 includes analyzing / searching the video stream starting from the current frame and progressing backward in time during the predetermined time period 120. If no prior event is found, a visual indicator 116 can be added S1008. If a prior event is found, the analysis / search can be terminated early to conserve computer processing resources. Since a prior event has already been found in this case, the step of adding a visual indicator should not be performed.

[0072] Additionally, a prior event may refer to an event that occurs continuously or intermittently in a video stream. The prior event may have been initially detected in a manner similar to other events, such as new event 112. The predetermined time period may refer to a time or a corresponding number of subsequent frames. The predetermined time period may preferably be in the range of 1-3600 seconds, more preferably in the range of 5-300 seconds, and most preferably in the range of 5-60 seconds.

[0073] The wording whether a prior event has been detected within the new event area 114 may alternatively be understood as whether a prior event corresponding to the new event area 114 has been detected. This correspondence may, for example, be understood as the spatial extent (within the image frame 103) of the prior event and the new event area 114 at least partially overlapping. It may also be understood as the prior event and the new event area 114 substantially corresponding to each other with respect to position, size, shape, or any combination of the foregoing. For example, a tiny new event detected within or overlapping a larger area prior event may still be important or relevant. Indicating such events may prevent them from being obscured by larger events in the video stream. The prior event in the scene may be detected by the same video capture device that captured the new event 112 or by another video capture device.

[0074] When no prior event is detected during the predetermined time period, the method further includes adding (S1008) a visual indicator 116 to the image frame 103 corresponding to the detection time of the new event 112 and to a plurality of subsequent image frames 103 of the video stream. The image frame 103 to which the visual indicator 116 is added (S1008), and subsequent image frames 103, may include the new event 112.

[0075] Visual indicator 116 may coincide with new event area 114. Visual indicator 116 may correspond in size or position to new event 112 in the video stream.

[0076] The visual indicator 116 may be configured to look like, for example, a crosshair or a marker overlaid onto the image frame 103 of the video stream. Figure 2b to Figure 2d The figure shows that the visual indicator 116 is configured as a crosshair. Various alternative configurations of the visual indicator 116 are available. The visual indicator 116 can be configured as essentially any geometric shape or object. The visual indicator 116 can include text or characters in certain circumstances. The visual indicator 116 can be characterized by an appearance that is configured to gradually change throughout multiple subsequent image frames 103.

[0077] The visual indicator 116 can be configured to have a visible appearance that draws the visual attention of an operator viewing the video stream to the new event 112 indicated by the visual indicator 116. The visual indicator 116 can be configured to not obstruct viewing or hide the indicated new event 112. Some overlap between the visual indicator 116 and the new event 112 may be acceptable. The image frame 103 (or video stream) can be configured with an exclusion zone in which no visual indicator is added.

[0078] Visual indicator 116 can be monochromatic or multi-colored. Visual indicator 116 can be characterized by only one absolute color, such as 100% black (RGB 0, 0, 0) or 100% white (RGB 255, 255, 255). Visual indicator 116 can be colored to provide a sufficient level of contrast with respect to the current background of image frame 103, specifically with respect to location in the current background area of ​​new event 112. Visual indicator 116 can be characterized by a flashing behavior between subsequent image frames, creating, for example, a strobing appearance for particularly important and urgent new event 112.

[0079] The step of adding S1008 the visual indicator 116 comprises gradually changing the appearance of the visual indicator 116 throughout a plurality of subsequent image frames 103 .

[0080] Gradually changing the appearance of the visual indicator 116 may include one or more of the following: gradually expanding the visual indicator throughout a plurality of subsequent image frames 103, gradually fading the visual indicator out (i.e., making it transparent), gradually changing the form of the visual indicator, gradually rotating the visual indicator, and gradually changing the color of the visual indicator 116. Gradually changing the appearance of the visual indicator 116 may further include one or more of the following: gradually making its line thinner or thicker, gradually shrinking the visual indicator, and gradually fading in the visual indicator.

[0081] Gradually changing can be understood as slightly changing the appearance of visual indicator 116 for each image frame 103 relative to the preceding image frame. It can also be understood as changing the appearance of visual indicator 116 in steps corresponding to a plurality of image frames 103. Generally, the step of gradually changing the appearance of visual indicator 116 means changing the appearance in a consistent manner across a plurality of subsequent image frames 103. Gradually changing the appearance of visual indicator 116 can include changing, for example, the color or transparency value of a pixel according to a linear, exponential, or logarithmic function.

[0082] An example of gradually changing the appearance of visual indicator 116 by fading it out may be understood as visual indicator 116 being 0% transparent in a first image frame, visual indicator 116 being 50% transparent in a fifth image frame, and visual indicator 116 being 100% transparent in a tenth image frame.

[0083] The method further includes displaying S1010 the video stream with the added visual indicator 116 on the display 101 of the client device. In general, the steps of the present disclosure can be at least partially performed by a device with processing capabilities (e.g., a client device). The display step S1010 can be performed by the display 101 of the client device.

[0084] The video stream may be one of a plurality of video streams. The display 101 may be adapted to view the plurality of video streams. The display 101 may be one of a plurality of displays. The display 101 may be a display viewed or intended to be viewed by a professional surveillance camera operator.

[0085] As Figure 1 As indicated by the optional step (with a dashed border) in , different embodiments may further include determining S2001 a duration from the last (i.e., most recent) prior event in new event area 114. The number of subsequent image frames 103 and / or the appearance of visual indicator 116 may be based on the determined duration.

[0086] For example, if the time period since the last prior event is determined to be long, visual indicator 116 may be maintained in the video stream for a greater number of subsequent image frames 103. Additionally or alternatively, the appearance of visual indicator 116 may be adapted to be more apparent to an operator viewing the video stream.

[0087] In-depth Figures 2a to 2d , the illustrated image frames 103a to 103d are image frames of a video stream ordered in time sequence, wherein Figure 2a is the first image frame 103a and Figure 2d This is the final image frame 103d. Figures 2a to 2d The image frames 103a to 103d in the example are not necessarily directly following or chronologically adjacent to each other in the video stream. Figure 2a The image frame 103a, and the events it depicts, can be understood as being in the video stream at Figure 2b 1 second or 30 image frames before the image frame 103b of .This can also be the case for the other figures herein that depict image frames 103 at discrete points in time. Figure 2a An image frame 103a is shown depicting a scene 100 including a house, a tree, and a person. Figures 2a to 2d and the bold arrows in the following figures indicate events of motion in the scene. This is demonstrated by smoke rising from the chimneys of houses into the sky, by leaves and branches of trees swaying in the wind, and by people moving their arms up and down. It should be noted that all events in the scene (including those mentioned above) can be detected as new events, thus also facilitating the determination of new event regions. However, visual indicators are only added after it is determined or assumed that no previous events have been detected during a previous predetermined time period. Previous events may precede all of these events because they are in progress and therefore, in Figure 2a No visual indicator is added.

[0088] Figure 2b Shown depicted in Figure 2a The image frame 103b of the scene 100 at a later point in time than the image frame 103a of the scene 100 is compared to the image frame 103b of the scene 100 at a later point in time. The image frame 103b can be, but need not be Figure 2a The image frame in the image frame immediately following the image frame 103a. Figure 2b In FIG, the door of the house has been opened. As disclosed, because of the motion associated with the opening of the door, the event is detected S1002 as a new event 112. The new event area 114 is determined S1004 as the dotted area containing the door. Figure 2bIn the example of , the check S1006 whether a previous event has been detected within the new event area 114 during a predetermined preceding time period is negative, and a visual indicator 116 is added S1008 to the image frame 103b.

[0089] Figure 2c Shown depicted in Figure 2b The image frame 103b is compared to the image frame 103c of the scene 100 at a later point in time. The significant change in the later image frame 103c is the visual indicator 116, which indicates that the image frame 103c has changed from the image frame 103b to the image frame 103c. Figure 2b The visual indicator 116 now appears as a thin dotted line instead of Figure 2b This can be understood as, for example, the visual indicator 116 gradually fading out. Figure 2c Another change is the appearance of a person behind the tree trunk. This can be understood as another new event 112 detected S1002. Another new event region 114 is also determined S1004. A check S1006 is performed to determine whether a previous event has been detected within new event region 114 during a predetermined preceding time period. This is again negative, and another visual indicator 116 is added S1008 to image frame 103c.

[0090] Figure 2d Shown depicted in Figure 2c The visual indicator 116 indicating that the person is behind the tree trunk has changed its appearance, as previously shown in FIG. Figure 2c In addition, since the door is closed again, a new event 112 is detected S1002. Another new event area 114 is determined S1004, but this time, no visual indicator is added. This is because the check S1006 whether a previous event has been detected in the new event area 114 during a predetermined previous time period is positive. This is because the same as in Figure 2b The events associated with the door being opened in Figure 2b The image frame 103b in Figure 2d The time between the image frames 103d is less than the predetermined time period. Figure 2b The door opened in Figure 2d The closed door in forms a preceding event during a predetermined preceding time period.

[0091] refer to Figure 3a to Figure 3b as well as Figure 4 , different embodiments of the method comprising additional steps will now be disclosed. Figure 3a to Figure 3b The diagram shows an example of how further steps may involve events and features of the image frame 103, while Figure 4A step-by-step demonstration of the methods of these embodiments is provided.

[0092] like Figure 4 As shown in , various embodiments may further include: detecting S3001 an ongoing discrete event 322 in scene 100 concurrently with new event 112. The discrete event 322 may be described as being similar to the new event or to an event already generally described above. More than one discrete event 322 may be detected S3001 within a single image frame 103. Upon detecting the discrete event 322, the method may further include: determining S3003 a discrete event region 324 in scene 100 within which the discrete event 322 was detected. More than one discrete event region 114 may be determined S3003 within a single image frame 103.

[0093] A discrete event region may, for example, correspond to an area outlined by pixels associated with discrete event 322 that have moved between image frames 103. Discrete event region 324 may alternatively correspond to an arbitrary geometric shape, not necessarily corresponding in size or position to discrete event 322. Discrete event region 324 may correspond to previous new event region 114.

[0094] The method may further include determining S3005 a distance 330 between the new event region 114 and the discrete event region 324 .

[0095] Distance 330 may correspond to a distance within the planar image frame 103 depiction of scene 100. Distance 330 may be determined, for example, by Figure 3a to Figure 3b An example of distance 330 is shown as the closest distance between the boundaries of event areas 114, 324. Distance 330 may alternatively be determined as the distance between the center points of event areas 114, 324. Distance 330 may alternatively be determined as the farthest distance between the boundaries of event areas 114, 324, or a combination of the above examples. Distance 330 may be determined in the coordinate system of image frame 103.

[0096] Distance 330 may alternatively correspond to a predicted or estimated real-world distance between events in scene 100, i.e., a distance in actual scene 100. Such an estimate may provide a depth aspect to distance 330, which may be advantageous because the focus setting of a video capture device may affect the quality of depiction of features in scene 100 differently depending on their distance from the video capture device. Distance 330 may be determined, for example, by computer image analysis or by an external sensor suitable for measuring distance, such as a radar device.

[0097] Upon determining that the distance 330 between the new event area 114 and the discrete event area 324 is greater than the predetermined threshold, the step of adding S1008 the visual indicator 116 may be performed.

[0098] Alternatively, upon determining that distance 330 between new event area 114 and discrete event area 324 is less than a predetermined threshold, add S1008 visual indicator 116. Such an embodiment may alleviate the problem of discrete events 322 obscuring new events 112 that are close to or overlap discrete event area 322.

[0099] The number of subsequent image frames 103 and / or the appearance of visual indicator 116 may be based on distance 330 between new event area 114 and discrete event area 324 .

[0100] For example, if the distance to the closest discrete event 322 is determined to be longer, the visual indicator 116 may be maintained in the video stream for a greater number of subsequent image frames 103. Additionally or alternatively, the appearance of the visual indicator 116 may be adapted to be more apparent to an operator viewing the video stream.

[0101] A wide range of different image frame 103 sizes are possible. In this regard, the predetermined threshold can be considered as a percentage of the diagonal distance of the image frame 103. Preferably, the predetermined threshold for the distance is in the range of 1-50%, more preferably the predetermined threshold for the distance is in the range of 5-30%, and most preferably the predetermined threshold for the distance is in the range of 10-20%.

[0102] Figure 3a to Figure 3b It is shown that the rising smoke from the chimney, leaves and branches, and the moving arms of a person can all be detected S3001 as separate events 322. Separate event regions 324 have been determined S3003 for each of the separate events. The opening of a house door is detected S1002 as a new event 112, and a new event region 114 surrounding it is determined S1004.

[0103] exist Figure 3a , two distances 330a and 330b are shown. First distance 330a is the distance between discrete event area 324 corresponding to the person's moving arm and new event area 114 for the door. Distance 330a is less than a predetermined threshold. Second distance 330b is the distance between discrete event area 324 corresponding to the tree's leaf and branch movement and new event area 114 for the door. Distance 330b is greater than the predetermined threshold. Because one of the distances (i.e., the first distance) is less than the predetermined threshold, no visual indicator is added.

[0104] exist Figure 3b, two distances 330c and 330d are shown. The first distance 330c is the distance between the discrete event area 324 corresponding to the person's moving arm and the new event area 114 for the door. Because the person is now further away from the door, this distance 330c is greater than a predetermined threshold. The second distance 330d is the distance between the discrete event area 324 corresponding to the tree's leaves and branches moving and the new event area 114 for the door. Figure 3a The same as the distance 330b in the case of , the distance 330d is still greater than the predetermined threshold. Since both the distances 330c and 330d are greater than the predetermined threshold, the visual indicator 116 is added S1008 to the image frame 103.

[0105] It should be noted that the above disclosure is from the perspective that the new event 112 is the opening of the door in the scene 100. Figure 3a to Figure 3b , and other concurrently occurring events form discrete events 322. However, each discrete event 322 also forms a new event that is evaluated in the same manner as new event 112, and may be provided with a visual indicator if the condition in question is met.

[0106] about Figure 5a to Figure 5b as well as Figure 6 , different embodiments of the method will now be disclosed including additional steps regarding establishing a motion level. Figure 5a to Figure 5b Provides examples illustrating how additional steps relate to events and features of image frame 103, while Figure 6 A distributed demonstration of the methods of these embodiments is provided.

[0107] like Figure 6 As shown in , different embodiments may further include determining S4001 a motion level of the new event 112. When it is determined that the motion level is greater than a predetermined threshold, the step of adding S1008 a visual indicator 116 may be performed.

[0108] The motion level may refer to a numerical value related to the motion detected or perceived in the new event region 114 for subsequent image frames 103. For example, the motion level may refer to the percentage of pixels within the new event region 114 that have been seen to have their color values ​​changed by more than a predetermined amount. The motion level may be determined using computer image analysis, motion detection, radar analysis, and the like.

[0109] Figure 5a to Figure 5b A simple example of the motion level filter in action is shown in FIG. Figure 5a, it is shown that a person's arm moving up and down in the scene 100 is detected S1002 as a new event 112, and a corresponding new event area 114 is determined S1004. However, no visual indicator is added because the motion level of the new event 112 determined S4001 is determined to be less than a predetermined threshold value regarding the motion level.

[0110] Figure 5b The same situation is shown, but in which the movement of the person's arm is substantially faster. Thus, the determined motion level S4001 is determined to be greater than the predetermined threshold for motion level. As such, visual indicators 116 are added S1008 to image frame 103 and subsequent image frames 103. The number of subsequent image frames 103 and / or the appearance of visual indicators 116 can be based on motion level in a manner similar to the duration and distance-based aspects disclosed above.

[0111] The predetermined threshold for the level of motion may be determined by a user / operator viewing the images.

[0112] about Figures 7a and 7b as well as Figure 8 , different embodiments of the method will now be disclosed which include further steps regarding the detection and size determination of objects. Figures 7a and 7b Examples are provided illustrating how these additional steps relate to events and features of image frame 103, while Figure 6 Provides a distributed demonstration of the embodiments.

[0113] like Figure 8 As shown in , different embodiments may further comprise: determining S5001 an object classification regarding an object, the object being associated with the new event 112. Adding S1008 a visual indicator may be performed based on the determined object, meaning that a visual indicator will be added for some predefined object types but not for others.

[0114] Figure 8 As also shown in FIG, different embodiments may further include determining S5003 the size of the object associated with the new event 112. The step of adding S1008 the visual indicator may be performed based on the determined size of the object, meaning that the visual indicator may be given a different appearance for objects of different sizes.

[0115] The number of subsequent image frames 103 and / or the appearance of the visual indicator 116 can be based on the determined object classification and / or the determined size of the object in a manner similar to the duration and distance-based aspects disclosed above. Overall, more significant, relevant, or potentially missed new events 112 can be provided with, for example, more prominent visual indicators 116 (which last longer and / or have a more prominent appearance), thereby further increasing the probability that the operator will direct their attention to the indicated area.

[0116] As previously defined, object classification refers to any group of objects that share a certain number of common features. Some examples of possible object classifications include humans, faces, vehicles, cars, smoke / fumes, fire, animals, birds, geometric shapes, etc. The size of an object can refer to the actual size of an object in the video stream or the predicted real-world size of an object. The object can be a physical object in scene 100, such as a door in a house, a person's arm, or the sky in the background. The size can be determined using computer image analysis, radar analysis, etc. The object classification can be determined using computer image analysis, object recognition, motion detection, radar analysis, etc. If image analysis is used, image frames 103 of the video stream can be used for the image analysis.

[0117] Object classification can be determined by an object detector or object classifier. The object classifier can be implemented as hardware designed or adapted for object detection. The object classifier can be implemented as computer instructions or software stored on a computer-readable storage medium that, when executed, instruct a computer or device with processing capabilities to perform object classification. Figure 7a An image frame 103 of a scene 100 similar to the previous figure is shown. However, here, a bird can be seen sticking its head out from the top of a tree. This corresponds to the detection S1002 of a new event 112, and the determination S1004 of a corresponding new event area 114. However, no visual indicator is added. This may be due to the determination S5001 that the object associated with the new event 112 is a bird and that the object belongs to the object classification of birds. In this example, birds may be defined as less important or relevant object classifications. This may be the case for a surveillance camera covering a parking lot, as it would not be necessary to warn a human operator every time a bird appears in scene 100.

[0118] An alternative example where birds might be considered as being classified as a high importance / relevance object might be a situation where a camera is configured to monitor an airport runway area. For this area of ​​application, detecting birds would be a high priority due to the hazard they could pose to aircraft landing and takeoff. Therefore, a visual indicator 116 could be added to a new event 112 associated with an object belonging to the bird object classification.

[0119] Figure 7b Shown with Figure 7a A similar image frame is shown in FIG, however, this time image frame 103 is provided with a visual indicator 116. This is the case even though the new event area 114 for the bird clearly overlaps with the event area for the tree. Figure 7bIn the case of adding S1008 a visual indicator 116, it can be a result of determining that the size of the object (i.e., the bird) associated with the new event 112 determined S5003 is smaller than, for example, a predetermined threshold. The predetermined threshold for size can, for example, be implemented in such a case as a ratio of the new event area 114 to the event area of ​​the tree. This can increase the chances of detecting small new events 112 or objects that would otherwise risk being obscured by larger objects or events. Although not demonstrated in the figures, the opposite situation is also possible, in which a visual indicator 116 is provided for larger new events 112 or objects and no visual indicator 116 is provided for smaller new events 112.

[0120] Figures 9a to 9b An image frame 103 of a video stream is shown illustrating how a visual indicator 116 may be configured to follow the movement of a new event 112 across subsequent image frames 103 of the video stream. Figure 9a In , a visual indicator 116 is added to the person appearing behind the tree. The person corresponds to a new event 112 detected S1002 with respect to the corresponding determined S1004 new event area 114. Figure 9b In the example, the visual indicator 116 is moved to follow the person's Figure 9a The visual indicator 116 may also be changed in appearance according to the above, and this is Figure 9a Demonstrated in.

[0121] about Figures 10a to 10b as well as Figure 11 , different embodiments of the method will now be disclosed including the additional step of ascertaining where the human operator is looking. Figures 10a to 10b An example of how these further steps relate to events and features of the image frame 103 is shown, while Figure 11 A distributed demonstration of the methods of these embodiments is provided.

[0122] like Figure 11 As shown in , different embodiments may further include determining S6001 a viewing point 640 of at least one eye of an operator viewing the display. When it is determined that the viewing point 640 is outside the display 101 or when it is determined that the distance 630 between the location of the viewing point 640 and the new event area 112 is greater than a predetermined threshold, the step of adding S1008 a visual indicator 116 may be performed.

[0123] Figure 10aAn image frame 103 similar to the previous image frame is shown. In image frame 103, two new events 112 are detected ( S1002 ), and corresponding new event areas 114 are determined ( S1004 ). However, a visual indicator 116 is added ( S1008 ) only to the new event 112 corresponding to the door of the house opening. No new event 112 corresponding to the person appearing behind the tree trunk is indicated. In this example, this is due to the distance 630 between the viewing point 640 and the corresponding new event area 114.

[0124] The distance 630a between the viewing point 640 and the new event area 114 for the door is greater than a predetermined threshold, while the distance 630b between the viewing point 640 and the new event area 114 for the person is less than the predetermined threshold. The reasoning behind this model is that new events 112 that are closer to the viewing point 640 determined S6001 are less likely to be missed by a human operator viewing the video stream.

[0125] Figure 10b and Figure 10a 103, in which only one new event 112—a person appearing behind a tree trunk—is presented. Viewpoint 640 is also repositioned to a point outside of image frame 103. In this case, distance 630 between viewpoint 640 and new event region 114 is less than a predetermined threshold; however, visual indicator 116 is still added (S1008) to new event 112. This is because viewpoint 640 is determined to be outside of image frame 103, which in this case corresponds to the viewpoint being outside of the display.

[0126] The viewing point 640 of at least one eye of the operator can be determined by gaze detection. The viewing point 640 can be determined using an eye tracking device. The eye tracking device can be integral to or communicatively connected to the display 101 on which the operator views the video stream. Gaze detection and / or eye tracking can be performed for multiple eyes of multiple operators viewing one or more video streams on one or more displays.

[0127] Tracking eyes can be achieved by including a camera pointed at the operator. Images obtained from such a camera can be used to determine the operator's viewing direction and viewing point 640 by known methods. Such a camera can be considered an eye tracking device.

[0128] When the viewing point 640 is determined to be outside of the display 101 or when the viewing point 640 cannot be fully determined (6001), the visual indicator 116 can be configured to attract the operator's attention by, for example, a strobing appearance or other similarly dramatic appearance. The method can be further adapted to trigger a sound from a connected speaker to gain the operator's attention. In this way, the operator's attention can be regained even if the operator averts their gaze, and in some cases, even if the operator has fallen asleep.

[0129] In general, the present disclosure discloses a number of different examples of filters for adding S1008 a visual indicator 116 to a new event 112. These filters can be combined with each other to produce a combined filter. A weight can be set for each filter in the combination, which generally scores the filter contribution of the resulting filter. It should be understood that the weighting can be set differently for different types of applications or environments.

[0130] Some of the filters described herein include adding S1008 visual indicators 116:

[0131] • when no prior event is detected during a predetermined time period;

[0132] • upon determining that the distance 330 between the new event region 114 and the discrete event region 324 is greater than a predetermined threshold;

[0133] ●When it is determined that the motion level is greater than a predetermined threshold;

[0134] Based on a defined object classification;

[0135] Based on the determined size of the object;

[0136] • when it is determined that the viewing point 640 is located outside of the display; and

[0137] • When it is determined that the viewing point 640 is located at a distance 630 from the new event area 114 greater than a predetermined threshold.

[0138] The present invention may also be implemented as a non-transitory computer-readable storage medium having instructions stored thereon that, when executed on a device having processing capabilities, are used to perform the computer-implemented methods discussed herein. The non-transitory computer-readable storage medium may be a data storage disk, such as a hard drive or an optical disk. The non-transitory computer-readable storage medium may be a solid-state memory device, such as a flash memory or a solid-state drive. The non-transitory computer-readable storage medium may be part of a client device.

[0139] Figure 12 The diagram shows a client device 701 for displaying a video stream of a scene 100. The client device 701 is shown connected to a video capture device 770 which captures the scene 100 and transmits it to the client device 701.

[0140] The client device 701 may be connected to multiple video capture devices 770. The client device 701 may be configured to receive multiple video streams from the multiple video capture devices 770. The client device 701 may include a decoder. The client device 701 may include a graphics processing unit. The client device 701 may include an encoder. Figure 12 As shown in , client device 701 can be included in a system with video capture device 770.

[0141] Video capture device 770 may be a video camera. Video capture device 770 may be suitable for monitoring or surveillance purposes. Video capture device 770 may be an integral part of client device 701 or physically disconnected from client device 701 and communicatively coupled to client device 701. Video capture device 770 may be configured to transmit a video stream to client device 701 wirelessly or by wire. Video capture device 770 may be part of a network of multiple connected video capture devices 770. Video capture device 770 may be a surveillance camera for wide-angle imaging, a surveillance camera for panoramic imaging, or a surveillance camera for 360-degree imaging.

[0142] Now consider Figure 12 Client device 701 includes control circuitry 750 configured to execute event detection functionality 752 configured to detect a new event in scene 100. Control circuitry 750 is further configured to execute event location functionality 754 configured to determine a new event region in scene 100 corresponding to a region within which the new event was detected.

[0143] The control circuit 750 is further configured to execute an event comparison function 756 configured to check whether a prior event has been detected within the new event region during a predetermined time period prior to the new event. The control circuit 750 is further configured to execute an indicator addition function 758 configured to add a visual indicator 116 to the image frame 103 and to a plurality of subsequent image frames 103 of the video stream when a prior event has not been detected during the predetermined time period. In this embodiment, the visual indicator 116 coincides with the new event region.

[0144] The indicator adding functionality 758 is further configured to gradually change the appearance of the visual indicator 116 throughout a plurality of image frames following the currently displayed image frame 103. The control circuitry 750 is further configured to execute a display functionality 760 configured to display the video stream with the added visual indicator 116 on the display 101 of the client device 701.

[0145] The display 101 may be an integral part of the client device 701 or physically disconnected from the client device 701 as well as communicatively coupled to the client device 701 .

[0146] Additionally, variations to the disclosed embodiments can be understood and effected by those skilled in the art in practicing the claimed invention, from a study of the drawings, the disclosure, and the appended claims.

Claims

1. A computer-implemented method for displaying a video stream of a scene captured by a video capture device on a display of a client device, the method comprising: detecting new events in the scene, determining a new event region in the scene within which the new event is detected, checking whether a previous event has been detected within the new event area during a predetermined time period prior to the new event, detecting an ongoing discrete event in the scene concurrently with the new event, Upon detecting the discrete event, determining a discrete event region in the scene within which the discrete event is detected, determining a distance between the new event area and the discrete event area, and upon no prior event being detected during the predetermined time period and upon determining that the distance between the new event area and the discrete event area is greater than a predetermined threshold, adding a visual indicator to an image frame corresponding to a time of detection of the new event and to a plurality of subsequent image frames of the video stream, wherein the visual indicator coincides with the new event area, and wherein adding the visual indicator comprises gradually changing an appearance of the visual indicator throughout the plurality of subsequent image frames, and The video stream with the added visual indicator is displayed on the display of the client device.

2. The computer-implemented method of claim 1 , further comprising: A duration from a last preceding event in the new event region is determined, wherein a number of subsequent image frames and / or an appearance of the visual indicator is based on the duration.

3. The computer-implemented method of claim 1 , wherein: The number of subsequent image frames and / or the appearance of the visual indicator is based on the distance between the new event area and the discrete event area.

4. The computer-implemented method of claim 1 , comprising: A motion level of the new event is determined, wherein the step of adding the visual indicator is performed upon determining that the motion level is greater than a predetermined threshold.

5. The computer-implemented method of claim 1 , comprising: A motion level of the new event is determined, wherein a number of subsequent image frames and / or an appearance of the visual indicator is based on the motion level.

6. The computer-implemented method of claim 1 , further comprising: An object classification is determined for an object associated with the new event, wherein the step of adding the visual indicator is performed based on any determined object classification.

7. The computer-implemented method of claim 1 , further comprising: A size of an object associated with the new event is determined, wherein the step of adding the visual indicator is performed based on the determined size of the object.

8. The computer-implemented method of claim 1 , wherein: The visual indicator is configured to follow movement of the new event across subsequent image frames of the video stream.

9. The computer-implemented method of claim 1 , wherein: The step of gradually changing the appearance of the visual indicator includes one or more of: gradually expanding the visual indicator throughout the multiple subsequent image frames, gradually fading the visual indicator, gradually changing the form of the visual indicator, gradually rotating the visual indicator, and gradually changing the color of the visual indicator.

10. The computer-implemented method of claim 1 , further comprising: determining a viewing point of at least one eye of an operator viewing the display, Wherein, the step of adding the visual indicator is performed when it is determined that the viewing point is outside the display or when it is determined that the distance between the viewing point and the new event area is greater than a predetermined threshold.

11. A non-transitory computer-readable storage medium having stored thereon instructions for implementing the computer-implemented method of claim 1 when executed on a device having processing capabilities.

12. A client device for displaying a video stream of a scene, the client device comprising a control circuit configured to perform the following steps: an event detection function configured to detect new events in the scene and, concurrently with the new events, detect ongoing discrete events in the scene, an event location function configured to determine a new event region in the scene, the new event region corresponding to a region in the scene within which the new event was detected, and to determine a discrete event region in the scene within which the discrete event was detected, an event comparison function configured to check whether a previous event has been detected within the new event area during a predetermined time period before the new event and to determine a distance between the new event area and the discrete event area, an indicator adding function configured to add a visual indicator to an image frame and to a plurality of subsequent image frames of the video stream when no prior event is detected during the predetermined time period and when the distance between the new event area and the discrete event area is determined to be greater than a predetermined threshold, wherein The visual indicator is consistent with the new event area, and wherein the indicator adding functionality is further configured to gradually change the appearance of the visual indicator throughout the plurality of subsequent image frames, and A display function is configured to display the video stream with the added visual indicator on a display of the client device.

13. A system comprising: The client device according to claim 12, and Video capture device.

14. The system according to claim 13, wherein: The video capturing device is a surveillance camera for wide-angle imaging, a surveillance camera for panoramic imaging, or a surveillance camera for 360-degree imaging.

Images