System for simulating depth of field, portable device for generating a depth-of-field effect and computer program product

Abstract

A system for simulating depth of field, comprising: a) a display device; b) a live camera configured to capture a video signal of a scene; c) a sensor configured to capture eye data of a viewer, including their gaze direction; and d) a processing unit configured to: i) determine a focal plane within the scene from the captured eye data; and ii) dynamically control the focal plane and / or aperture of the live camera based on the determined eye data, such that the image captured by the camera and displayed on the display device has a depth-of-field effect corresponding to natural vision.

Description

The present invention relates to systems, portable devices, and computer program products for simulating depth of field when displaying images on display devices. In particular, the invention relates to technologies that, by capturing eye data from a viewer, enable a dynamic adjustment of the displayed image or the perception of the image in order to create a realistic depth-of-field effect and thus allow for intuitive distance estimation. The human ability to perceive spatial depth and estimate distances is based on a combination of several visual mechanisms. Among the most important are stereoscopic perception through two eyes (binocular disparity), the interpretation of size relationships and occlusion of objects (monocular cues), and the dynamic adjustment of the eye's focus. This third mechanism, which relies on the function of the eye's lens and iris, creates what is known as depth of field. When a person focuses on an object at a certain distance, the eye's lens adjusts its curvature to focus that object sharply onto the retina. Objects located in front of or behind this focal plane appear increasingly blurred the greater their distance from it. By consciously or unconsciously changing focus, humans can thus estimate the relative distances of objects in their field of vision. When viewing conventional two-dimensional displays, such as monitors or screens, this natural mechanism of depth perception is lost. The displayed image is a flat projection of a three-dimensional scene, where the depth of field is already fixed at the time of capture by the camera settings (focal length and aperture). The viewer's eye always focuses on the physical plane of the screen itself and not on the virtual objects within the displayed scene. A dynamic adjustment of the eye's focus for distance estimation is therefore not possible. The main disadvantage of the current state of the art is that the inability to utilize the eye's natural focusing mechanism leads to unreliable and often purely cognitive distance estimation. While a fixed, shallow depth of field is deliberately used in artistic media such as film or photography to direct the viewer's attention, this effect is detrimental and even dangerous in safety-critical or immersive applications. For example, when using a digital rearview mirror, a driver may find it difficult to intuitively judge the distance and speed of following vehicles because the image has a fixed focal plane. Similar problems arise with the remote control of drones and robots, or in medical endoscopy applications, where precise depth perception is crucial for the safe and accurate execution of tasks. In virtual worlds and video games, immersion is also limited by the inability to naturally "capture" the scene. The viewer is forced to accept the system's predetermined focus instead of deciding for themselves which part of the scene they want to see in sharp focus. The object of the present invention is therefore to overcome the disadvantages of the prior art and to provide a system, a device, and a computer program product that simulate the natural depth perception of the human eye when viewing two-dimensional display devices. In particular, an intuitive and realistic distance estimation of objects in a displayed scene should be made possible by generating a dynamic depth-of-field effect that reacts to the viewer's viewing direction and focus. The aforementioned problem is solved by the features of independent claims 1, 7, 13, 20 and 22. The dependent claims represent advantageous embodiments of the invention. Claim 1 discloses a system for simulating depth of field based on the dynamic control of a live camera. The system comprises a display device, a live camera, a sensor for capturing eye data, and a processing unit. The core concept of this embodiment is to adapt the parameters of the recording camera to the viewer's intention in real time. The sensor detects where the viewer is looking, and the processing unit translates this information into a control command for the camera. The processing unit uses the eye data to determine a focal plane within the scene captured by the camera. Based on this determination, it controls the focus plane of the live camera, for example, by activating the autofocus motor. Additionally or alternatively, it can control the camera's aperture. A smaller aperture results in a greater depth of field, while a larger aperture enhances the depth-of-field effect. This dynamic coupling between the viewer's eye and the camera manipulates the image displayed on the screen to correspond to what the viewer would see if they were looking directly at the scene. The technical advantage of this approach is the direct and physically accurate generation of the depth-of-field effect at the source. No complex post-processing algorithms are needed to artificially create blur. This results in very low latency between the viewer's change in perspective and the image adjustment, which is essential for real-time applications such as digital rearview mirrors or robot control. The result is an extremely realistic and intuitive visual feedback that significantly increases safety and ease of use. As an alternative implementation, the processing unit could adjust not only the focus plane but also the focal length (zoom) of the camera to further naturalize the visual impression. According to a preferred embodiment, as described in claim 2, the sensor additionally detects the curvature of the eye lens. The processing unit is then configured to calculate the focal plane directly from this curvature. This represents a significant improvement over simply detecting the gaze direction. While the gaze direction only indicates a point on a 2D plane in the image, the curvature of the eye lens directly represents the physiological effort of the eye to focus on a specific distance. This approach is significantly more accurate and robust because it captures the viewer's actual focusing intention. It can resolve ambiguities that arise when multiple objects lie along the same line of sight. For example, the viewer might focus on a more distant object through a grid; the viewing direction would be similar for both objects, but the lens curvature is unambiguous. The resulting simulation is therefore more precise and responds more accurately to the user's intention, leading to a more natural and less tiring interaction. Claim 3 describes an alternative or complementary method for determining the camera focus plane, wherein the processing unit utilizes an autofocus function of the camera. After the processing unit has identified a target object or target area from the viewer's eye data, it can instruct the camera's autofocus to focus precisely on this area. This design has the advantage of leveraging established and highly optimized autofocus systems in modern cameras. Instead of calculating the focus plane itself and directly controlling the lens motor, the processing unit delegates this task to the camera hardware. This simplifies the implementation of the processing unit and utilizes the often superior speed and precision of phase-detection or contrast-detection autofocus systems. In an alternative design, the processing unit could specify several potential target areas to the autofocus system, weighted according to the probability of the user's intention, and the camera would then select the optimal focus point. According to claim 4, the camera's aperture is dynamically adjusted to the detected pupil size of the viewer. The pupil of the human eye dilates in low light and constricts in bright light, which directly affects the depth of field. By mimicking this physiological response, the system makes the simulation even more realistic. The main advantage lies in the consistent visual experience. For example, when the viewer looks from a bright to a dark scene, their pupil dilates, and they expect a shallower depth of field. The system simulates precisely this effect by also opening the camera aperture. This prevents visual dissonance and enhances the sense of presence and realism. Furthermore, this adjustment can help to optimally regulate the amount of light hitting the camera sensor, which improves image quality. Claim 5 proposes a simplified yet effective variant in which the camera's aperture is set to a fixed, static value. This value is chosen such that the aperture is large enough to produce a clearly perceptible depth-of-field gradient, i.e., a relatively shallow depth of field. The advantage of this design lies in its simplicity and robustness. Dynamic aperture control is not required, reducing the system's complexity and cost. Although the aperture is not adjusted to the viewer's pupil size, the core functionality—dynamic shifting of the focal plane based on gaze—is retained. This variant is particularly suitable for applications where cost is a critical factor or where lighting conditions are relatively constant, making dynamic aperture adjustment offer no significant added value. The applications mentioned in claim 6 – digital rearview mirror, drone / robot control, medical endoscope – particularly benefit from the invention. In all these scenarios, the operator is physically separated from the scene and relies on the display on a screen. The ability to intuitively and quickly assess distances is therefore not only a matter of convenience, but also of safety and effectiveness. A surgeon using an endoscope can better assess the spatial arrangement of tissues and instruments by simulating depth of field, thus increasing precision and reducing the risk of errors. A drone pilot can detect obstacles more reliably and estimate the distance to them more accurately. A driver can assess the traffic behind them more safely. The invention transforms the display device from a passive information medium into an active tool that supports and enhances the natural capabilities of human vision. Claim 7 discloses a second principal embodiment of the invention, which is based on the post-processing of a digital image. Instead of controlling a live camera, this system starts from an image source that already provides a digital image with associated depth information (e.g., a depth map). The processing unit modifies this image by applying a degree of blur that depends on the distance to the focal plane. This embodiment is extremely flexible because it does not rely on a live camera. It can be applied to stored photos, videos, or, as described in claim 9, to computer-generated scenes from video games or VR applications. The key advantage is the decoupling of recording and display. Ideally, the original image is captured or rendered with a very high depth of field (everything is in focus). The depth of field is artificially created only afterward. This allows for complete control over the visual effect. The processing unit can precisely calculate and apply the degree of blur to create an aesthetically pleasing or physically accurate "bokeh" effect. Furthermore, this approach is ideal for applications where no live camera is available or where controlling it is not possible or desired. As an alternative to applying a degree of blur, the processing unit could also simulate other effects such as chromatic aberration or vignetting, which are also associated with a shallow depth of field. Claim 8 describes a specific method for obtaining depth information: the combination of images from multiple cameras with different focal planes. This stereoscopic or multifocal approach is a robust method for creating an accurate depth map for a scene. The advantage lies in the high quality of the generated depth information. By comparing images from two or more cameras, the system can determine the distance to any point in the scene very accurately through triangulation or by analyzing focus differences. This leads to a more precise and artifact-free simulation of depth of field, as the transitions between sharp and blurred areas are based on a reliable 3D reconstruction of the scene. This method is particularly advantageous in complex scenes with many overlapping objects. Claim 13 defines a generalized system that summarizes the core features of the post-processing implementation. It emphasizes the functional chain: acquiring eye data, providing an image with depth information, determining the focus plane, modifying the image, and displaying the modified image. This claim serves as a generic term for the further developments described in claims 14 to 19. Claim 20 discloses a third, particularly innovative principal embodiment: a wearable device, for example in the form of glasses. Instead of altering the image displayed on the screen, this device directly modifies the viewer's perception. It does this by means of an optical element with controllable transmittance or translucency, which is arranged in the user's field of vision. The revolutionary advantage of this approach is that it works with any external display device without requiring modification, and multiple viewers can use the effect simultaneously on the same output device. The eye-tracking sensor is integrated into the glasses. The glasses receive depth information about the displayed image (e.g., wirelessly from the display device). The processing unit in the glasses determines the focal plane and then controls the optical element—for example, an LCD layer as specified in claim 21—such that the areas of the field of view corresponding to objects outside the focal plane are blurred or made translucent. The viewer sees a sharp image on the screen, but the glasses "paint" the blur directly into their field of vision. This allows for a universal retrofit of depth-of-field simulation for existing monitors, televisions, or even cinema screens. Finally, claim 22 discloses a computer program product that contains the instructions for performing the functions of the described systems and devices. This protects the software underlying the invention, which may be stored on a machine-readable medium. The advantage lies in protecting the actual intelligence of the system, which is implemented in the software. This enables the invention to be marketed as a pure software upgrade for existing hardware (e.g., for VR headsets or camera systems with eye-tracking functionality) or as a standalone product. The invention can also be described as a method for simulating depth of field, which is carried out by the systems mentioned above. The method comprises the computer-implemented steps of acquiring eye data from a viewer, providing an image with associated depth information, determining a focal plane based on the eye data, modifying the image to apply a dependent degree of blur, and displaying the modified image. As already discussed, this method finds advantageous application in a digital rearview mirror, in the control of drones or robots, in endoscopes, or in video game applications, where it significantly increases safety, precision, and immersion. According to a preferred embodiment, the sensor may further be configured to detect the curvature of the viewer's eye lens, and the processing unit may be configured to calculate the focal plane from the curvature of the eye lens. This allows for particularly precise and direct detection of the user's focusing intention, which increases the accuracy of the simulation. The controls feel more natural and responsive as a result, since they are linked to a direct physiological action. Furthermore, ambiguities regarding objects in the same line of sight are effectively resolved. An alternative development approach involves configuring the processing unit to determine the camera's focus plane using the camera's autofocus function. This simplifies implementation by leveraging the camera's highly optimized and fast autofocus hardware. This can reduce system latency and increase focusing robustness. Simultaneously, it lowers the computational demands on the external processing unit. According to implementations of the invention, the sensor is further configured to detect the pupil size of the viewer, and the processing unit is configured to dynamically adjust the aperture of the camera to the detected pupil size. The advantage of this design is increased realism in the simulation, as the physical properties of human vision are replicated. Adjusting the depth of field to the ambient brightness (via pupil size) creates a consistent visual experience. This prevents visual dissonance and can also improve image quality through optimized light control. According to a preferred embodiment, the camera's aperture is set to a static value large enough to produce a perceptible depth of field gradient. This variant significantly reduces the technical complexity and cost of the system, as there is no need to control movable aperture blades. It offers a simple and robust solution that retains the core functionality of dynamic focus shifting. This makes the technology more accessible to price-sensitive mass markets. According to a preferred embodiment, the image source obtains depth information by combining images from multiple cameras with different focal planes. This method results in highly accurate and detailed depth maps, which significantly improves the quality of the blur simulation. Artifacts at the edges of objects are minimized, and the transitions between sharp and blurred areas appear smoother and more natural. This enhances the visual realism and overall quality of the generated image. According to implementations of the invention, the image source is a computer-generated scene from a video game or a virtual reality application. This application enables a completely new level of immersion in digital worlds. Players can naturally "scan" the virtual environment with their eyes, greatly enhancing the feeling of presence and realism. Furthermore, game developers can use this feature to create new interaction mechanics based on the player's focus. In an alternative development, the processing unit could be configured to adjust the degree of blur based on the ambient lighting conditions or to allow manual manipulation. This adaptation to ambient light conditions ensures a consistent perception of depth of field, similar to dynamic aperture control. The ability to manually manipulate the system gives the user or developer creative control to enhance the effect for artistic purposes or to weaken it for analytical purposes. This increases the system's flexibility and range of applications. According to a preferred embodiment, the processing unit is further configured to take into account the viewer's distance from the display device and / or the display device's resolution when calculating the degree of blur. This results in a more physically accurate and visually consistent simulation. The perceived blur depends on these factors, and considering them ensures that the effect appears consistently realistic regardless of the viewing situation. This increases the robustness and quality of the system across different hardware setups and usage scenarios. According to a preferred embodiment, the sensor is further configured to detect the pupil size of the viewer, and the processing unit is configured to scale the applied blur level based on the detected pupil size. This technique simulates the function of the iris in post-processing, significantly increasing realism. It mimics how the depth of field of the human eye changes under different lighting conditions. This creates a coherent and believable visual experience for the user. According to implementations of the invention, the optical element of the portable device according to claim 20 comprises a liquid crystal layer (LCD layer) whose transparency can be controlled separately for individual areas of the field of view. The use of a liquid crystal layer is technologically mature and cost-effective to manufacture. It enables very precise, pixel-accurate control of transparency, allowing for the creation of smooth and natural blur gradients. Furthermore, the switching speed of LCDs is high enough to ensure low latency, making the response to changes in viewing direction immediate and preventing discomfort. According to a preferred embodiment, in a system according to claim 7 or 8, the image source may be a computer-generated scene from a video game or a virtual reality application. This application in interactive media creates an unprecedented level of immersion. The player can explore the virtual world with their eyes just as they would the real world, greatly enhancing the feeling of presence and realism. This also gives game developers the opportunity to design innovative game mechanics that respond directly to the player's focus. In an alternative further development, it may be provided that in a system according to one of claims 7 to 9, the processing unit is further configured to adjust the degree of blur based on the ambient light conditions or to make it manually manipulable. Automatic adjustment to ambient light conditions ensures a consistent and realistic perception of depth of field, corresponding to natural vision. The option for manual manipulation, on the other hand, offers users or developers creative freedom to emphasize the effect for artistic purposes or reduce it for analytical purposes. This significantly increases the system's flexibility and range of applications. According to implementations of the invention, in a system according to any one of claims 7 to 10, the processing unit is further configured to take into account the distance of the viewer to the display device and / or the resolution of the display device when calculating the degree of blur. This leads to a physically more accurate and therefore visually more convincing simulation. Since the perceived blurriness depends on these parameters, taking them into account ensures that the effect remains consistent and realistic across different hardware configurations and viewing situations. This increases the robustness and perceived quality of the system. According to a preferred embodiment, in a system according to one of claims 7 to 11, the sensor is further configured to detect the pupil size of the viewer, and the processing unit is configured to scale the applied degree of blur based on the detected pupil size. This technique simulates the function of the iris in post-processing and significantly increases the realism of the rendering. It replicates how the depth of field of the human eye changes depending on the brightness. This creates a coherent and believable visual experience and avoids user irritation. According to a preferred embodiment, in a system according to claim 13, it may be provided that the eye data acquired by the sensor additionally include the curvature of the eye lens and / or the pupil size of the viewer. Capturing this additional physiological data allows for a significantly more direct and accurate determination of the user's focusing intention. Instead of deriving the intention solely from the direction of gaze, it is measured against a concrete physical action. This eliminates ambiguities and makes the system more responsive and intuitive to use. In an alternative further development, it may be provided that in a system according to claim 14, the processing unit is further configured to adjust the degree of blur based on the detected pupil size in order to simulate the opening of the viewer's iris. By simulating a fundamental physiological reaction of the eye, the visual experience becomes even more realistic. The user experiences a depth of field that adjusts to the perceived brightness, just like in natural vision. This increases immersion and prevents visual dissonances that the brain would perceive as unnatural. According to implementations of the invention, in a system according to one of claims 13 to 15, the image source is a live camera that provides a video signal and depth information in real time. This configuration enables the invention to be used in safety-critical, real-time applications such as digital vehicle mirrors or remote machine control. Low latency and direct coupling to a real-world scene are crucial here. The system acts as an extension of the operator's senses. According to a preferred embodiment, in a system according to claim 16, the processing unit is configured to dynamically control the focus plane of the live camera based on the captured eye data. This design creates a powerful hybrid system that combines the advantages of both approaches. It allows not only for the creation of a blur effect in post-processing, but also for the active control of the camera's physical focus. This enables an even more nuanced and physically accurate simulation of the visual impression. In an alternative development, a system according to one of claims 13 to 15 provides that the image source is a stored image, a film or a computer-generated scene and the depth information is in the form of a depth map. This approach offers maximum flexibility, as the depth-of-field effect can be applied to any existing content, provided a depth map is available. This opens up a wide range of applications in film post-production and the interactive playback of 3D content. Content creation is thus decoupled from interactive viewing. According to a preferred embodiment, in a system according to claim 18, the original image may have a high depth of field and the processing unit generates the degree of blur by subsequently applying an unsharp mask. This non-destructive workflow ensures that the maximum image information of the source material is preserved. The processing unit has full control over the characteristics of the blur, including the shape of the bokeh. This enables high-quality and artistically controllable creation of the depth-of-field effect. In summary, the present invention solves the fundamental problem that the natural depth perception of the human eye is lost when viewing two-dimensional screens. Normally, humans intuitively estimate distances by focusing on objects, which blurs other objects in the foreground and background. The invention artificially restores this mechanism by using an eye sensor to detect where the viewer is looking or attempting to focus. A processing unit uses this information to generate a dynamic depth-of-field image in real time, where only the targeted area appears sharp. The crucial advantage is that the viewer can once again use their natural visual ability to estimate distances, making interaction with displayed scenes safer and more immersive. A first key embodiment of the invention is aimed at real-time applications in which a live camera captures a scene. Here, the data acquired by the eye sensor is used to dynamically control the physical properties of the recording camera, such as its focus plane and aperture. When the viewer focuses their gaze on a distant object, the camera automatically focuses on that object. The advantage of this approach lies in its low latency and the physically accurate generation of the depth-of-field effect directly at the source. This is particularly beneficial for safety-critical applications such as digital vehicle mirrors, remote control of robots, or medical endoscopes, where immediate and realistic feedback is essential. A second, particularly flexible implementation is based on the digital post-processing of images. Instead of controlling a camera, the system modifies an existing digital image for which depth information, for example in the form of a depth map, is available. The processing unit analyzes the viewer's gaze data, determines the desired focal plane, and then artificially applies a blurring effect to all image areas that lie outside this plane. The great advantage of this method is its universality: it can be applied to any type of content, be it stored photos, films, or computer-generated scenes from video games and virtual worlds. This enables a completely new level of immersion and interactivity, as depth perception is no longer statically predetermined but controlled by the viewer. A third, particularly innovative embodiment shifts the generation of the effect away from the screen and directly to the viewer, achieved through a wearable device such as eyeglasses. These glasses contain the eye sensor and wirelessly receive the depth information of the image displayed on an external screen. A controllable optical element integrated into the user's field of vision, for example, a liquid crystal layer, is then controlled to selectively blur parts of the field of vision corresponding to objects outside the focal plane. The immense advantage of this concept lies in its universality and retrofittability; it works with any external display device without requiring any modifications, thus making the technology widely accessible. To further enhance the realism and precision of the simulation, the invention provides additional refinements. Instead of relying solely on the direction of gaze, the system can also detect the actual curvature of the eye's lens to determine the viewer's focusing intention even more accurately. Furthermore, the simulated depth-of-field effect can be dynamically adjusted to the viewer's pupil size as detected by the sensor. This mimics the eye's natural response to varying light conditions, resulting in a more visually consistent and less tiring experience, as the simulation takes the eye's physiological state into account. In summary, the invention bridges the gap between natural human perception and digital image display. It transforms passive screens into active visual tools that harmonize with the human visual system instead of counteracting it. The described embodiments open up wide-ranging application possibilities in the automotive industry, medical technology, robotics, and the entertainment industry, significantly increasing safety, precision, and immersion in each case. The core intelligence of the system can be implemented as a flexible computer program product, which allows for easy integration into existing or future hardware.

Claims

A system for simulating depth of field, comprising: a) a display device; b) a live camera configured to capture a video signal of a scene; c) a sensor configured to capture eye data of a viewer, including their gaze direction; and d) a processing unit configured to: i) determine a focal plane within the scene from the captured eye data; and ii) dynamically control the focal plane and / or aperture of the live camera based on the determined eye data, such that the image captured by the camera and displayed on the display device has a depth-of-field effect corresponding to natural vision. System according to claim 1, wherein the sensor is further configured to detect the curvature of the viewer's eye lens, and the processing unit is configured to calculate the focal plane from the curvature of the eye lens. System according to claim 1 or 2, wherein the processing unit is configured to determine the focus plane of the camera using an autofocus function of the camera. System according to one of the preceding claims, wherein the sensor is further configured to detect the pupil size of the viewer, and the processing unit is configured to dynamically adjust the aperture of the camera to the detected pupil size. System according to one of claims 1 to 3, wherein the aperture of the camera is set to a static value that is large enough to produce a perceptible sharpness gradient. System according to one of the preceding claims, wherein it is configured as a digital rearview mirror in a vehicle, for controlling a drone or robot, or as part of a medical endoscope. A system for simulating depth of field, comprising: a) a display device; b) an image source configured to provide a digital image with a high depth of field and associated depth information, in particular in the form of a depth map; c) a sensor configured to capture eye data of a viewer, including their gaze direction; and d) a processing unit configured to: i) determine a focal plane within the scene defined by the depth information from the captured eye data; and ii) modify the digital image by applying a degree of blur, dependent on the distance of image areas from the focal plane, to image areas outside the determined focal plane before displaying the modified image on the display device. System according to claim 7, wherein the image source obtains the depth information by combining images from multiple cameras with different focal planes. System according to claim 7 or 8, wherein the image source is a computer-generated scene from a video game or a virtual reality application. System according to one of claims 7 to 9, wherein the processing unit is further configured to adjust the degree of blur based on the ambient light conditions or to make it manually manipulable. System according to one of claims 7 to 10, wherein the processing unit is further configured to take into account the distance of the viewer to the display device and / or the resolution of the display device when calculating the degree of blur. System according to one of claims 7 to 11, wherein the sensor is further configured to detect the pupil size of the viewer, and the processing unit is configured to scale the applied blur level based on the detected pupil size. A system for simulating depth of field on a display device, comprising: a) a sensor for capturing eye data of a viewer, wherein the eye data includes at least the viewer's gaze direction; b) an image source providing an image with associated depth information defining the spatial arrangement of objects in a scene; c) a processing unit configured to: i) determine a focal plane within the scene defined by the depth information based on the captured eye data; and ii) modify the image provided by the image source such that image areas on the determined focal plane are displayed sharply and other image areas are displayed with a degree of blur depending on their distance from the focal plane; and d) the display device configured to display the modified image. System according to claim 13, wherein the eye data acquired by the sensor additionally includes the curvature of the eye lens and / or the pupil size of the viewer. System according to claim 14, wherein the processing unit is further configured to adjust the degree of blur based on the detected pupil size in order to simulate the opening of the viewer's iris. System according to one of claims 13 to 15, wherein the image source is a live camera that provides a video signal and depth information in real time. System according to claim 16, wherein the processing unit is configured to dynamically control the focus plane of the live camera based on the captured eye data. System according to one of claims 13 to 15, wherein the image source is a stored image, a film or a computer-generated scene and the depth information is in the form of a depth map. System according to claim 18, wherein the original image has a high depth of field and the processing unit generates the degree of blur by subsequently applying an unsharp mask. A portable device for generating a depth-of-field effect when viewing an external display device, comprising: a) a frame structure that can be worn on a user's head, in particular in the form of spectacles; b) a sensor integrated into the frame structure, configured to capture eye data of the user, including their gaze direction; c) an interface configured to receive depth information about an image displayed on the external display device; d) a processing unit connected to the sensor and the interface, configured to determine a focal plane from the eye data and the depth information;unde) an optical element arranged in the user's field of vision with controllable transmittance or translucency, which is controlled by the processing unit in such a way that it blurs the areas of the field of vision that correspond to objects outside the specified focal plane.; Portable device according to claim 20, wherein the optical element comprises a liquid crystal layer (LCD layer) whose transparency can be controlled separately for individual areas of the field of view. Computer program product stored on a machine-readable medium and containing instructions which, when executed by the processing unit of a system according to any one of claims 1 to 19 or a portable device according to claim 20 or 21, control their respective functions.