Abstract: Apparatuses, methods, and computer programs for compressing a neural network are disclosed. An apparatus includes at least one processor; and at least one non-transitory memory including computer program code, the memory and the computer program code configured to, with the at least one processor, cause the apparatus to: receive information from a second device, where the information comprises at least one parameter configured to be used for compression of a neural network, where the at least one parameter is in regard to at least one first aspect or task of the neural network; and compress the neural network, where the neural network is compressed based, at least partially, upon the at least one parameter received from the second device. The apparatus may also receive a compressed neural network from the second device, and further compress the compressed neural network based on the information.

Abstract: Optimization of a neural network, for example in a video codec at the decoder side, may be guided to limit overfitting. The encoder may encode video(s) with different qualities for different frames in the video. Low-quality frames may be used as both input and ground-truth during optimization. High-quality frames may be used to optimize the neural network so that higher-quality versions of lower-quality inputs may be predicted. The neural network may be trained to make such predictions by making a prediction based on a constructed low-quality input for which the corresponding high-quality version is known, comparing the prediction to the high-quality version, and fine-tuning the neural network to improve its ability to predict a high-quality version of a low-quality input. To limit overfitting, the neural network may be concurrently or in an alternating fashion trained with low-quality input for which a higher-quality version of the low-quality input is known.

Abstract: A method, an apparatus and a computer program are provided.

Abstract: According to an example aspect, there is provided an apparatus comprising at least one processing core, at least one memory including computer program code, the at least one memory and the computer program code being configured to, with the at least one processing core, cause the apparatus at least to obtain, using a first image recognition or depth sensing mechanism, first information defining an input space, obtain, using the first or a second image recognition or depth sensing mechanism, second information defining a virtual output area, and cause rendering of sensor information captured from the input space into the virtual output area.

Abstract: The invention relates to a method comprising receiving a set of input samples, said set of input images comprising real images and generated images; extracting a set of feature maps from multiple layers of a pre-trained neural network for both the real images and the generated images; determining statistics for each feature map of the set of feature maps; comparing statistics of the feature maps for the real images to statistics of the feature maps for the generated images by using a distance function to obtain a vector of distances; and averaging the distances of the vector of distances to have a value indicating a diversity of the generated images. The invention also relates to technical equipment for implementing the method.

Abstract: A method comprising: creating a visual indicator based on at least one of visual analysis or audio analysis performed for a content comprising at least one visual element, wherein the visual indicator is selectable such that upon a selection of the visual indicator, access to the content is provided.

Abstract: A method comprising: causing analysis of a portion of a visual scene; causing modification of a first sound object to modify a spatial extent of the first sound object in dependence upon the analysis of the portion of the visual scene corresponding to the first sound object; and causing rendering of the visual scene and the corresponding sound scene including of the modified first sound object with modified spatial extent.

Abstract: A method comprising: causing definition of a display window in a displayed virtual scene; displaying window content inside the display window, in the displayed virtual scene; in dependence upon a first user action, causing a first change in the window content displayed inside the display window to first window content different to the window content, without changing the display window; and in dependence upon a second user action causing a second change in the window content displayed in the display window to second window content, different to the first window content and the window content, and causing a variation in the display window to become a modified display window different to the display window.

Abstract: An apparatus configured to: •in respect of virtual reality content comprising video imagery configured to provide a virtual reality space for viewing in virtual reality, wherein a virtual reality view presented to a user provides for viewing of the virtual reality content, the virtual reality view comprising a spatial portion of the video imagery that forms the virtual reality space and being smaller in spatial extent than the spatial extent of the video imagery of the virtual reality space and •based on one or more of; •i) a viewing direction in the virtual reality space of at least one virtual reality view provided to the user; and •ii) a selected object in the video imagery, •providing for one or more of generation or display of causal summary content comprising selected content from the virtual reality content at least prior to a time point in the virtual reality content currently viewed by the user, the causal summary content at least focussed on an object or event appearing in the at least one virtual r

Abstract: A method, apparatus and computer program code is provided. The method comprises: causing display of a virtual object at a first position in virtual space, the virtual object having a visual position and an aural position at the first position; processing positional audio data based on the aural position of the virtual object being at the first position; causing positional audio to be output to a user based on the processed positional audio data; changing the aural position of the virtual object from the first position to a second position in the virtual space, while maintaining the visual position of the virtual object at the first position; further processing positional audio data based on the aural position of the virtual object being at the second position; and causing positional audio to be output to the user based on the further processed positional audio data, while maintaining the visual position of the virtual object at the first position.

Abstract: An apparatus for identifying which sound sources are associated with which microphone audio signals, the apparatus including a processor configured to: determine/receive a position/orientation of at least one sound source relative to a microphone array; receive at least one microphone audio signal, each microphone audio signal received from a microphone; receive an audio-focussed audio signal from the microphone array, wherein the audio-focussed audio signal is directed from the microphone array towards the one of the at least one sound source so as to enhance the audio-focussed audio signal; compare the audio-focussed audio signal against each microphone audio signal to identify a match between one of the at least one microphone audio signal and the audio focussed audio signal; and associate the one of the at least one microphone with the at least one sound source, based on the identified match.

Abstract: An apparatus caused to, in respect of determination of a direction towards an object in a space, the object having a tag capable of emitting a wireless signal for reception by a locator of a positioning system configured to determine the direction towards the tag based on the wireless signal, based on object orientation information comprising the orientation of the object as viewed by a camera capturing visual imagery of the space and predetermined tag position information comprising the position on the object where the tag is attached, provide for determination of at least whether or not the object is oriented such that the tag substantially faces the locator and thereby control of the positioning system in the determination of at least a viability of the wireless signal for determining the direction towards the object, the tag substantially facing the locator indicative of the viability of the wireless signal.

Abstract: A method comprising: rendering a first media scene based upon media content provided by a content-rendering application via one or more rendering devices worn by the user; determining a priority for an event that occurs near the user, the event being independent of the content-rendering application; and automatically modifying the rendered first media scene, to render a modified second media scene based at least in part upon media content provided by the content-rendering application and at least in part upon other media content associated with the event.

Abstract: A method, apparatus, and computer program product are provided for training a neural network or providing a pre-trained neural network with the weight-updates being compressible using at least a weight-update compression loss function and/or task loss function. The weight-update compression loss function can comprise a weight-update vector defined as a latest weight vector minus an initial weight vector before training. A pre-trained neural network can be compressed by pruning one or more small-valued weights. The training of the neural network can consider the compressibility of the neural network, for instance, using a compression loss function, such as a task loss and/or a weight-update compression loss. The compressed neural network can be applied within a decoding loop of an encoder side or in a post-processing stage, as well as at a decoder side.

Abstract: A method comprising: causing analysis of a portion of a visual scene; causing modification of a first sound object to modify a spatial extent of the first sound object in dependence upon the analysis of the portion of the visual scene corresponding to the first sound object; and causing rendering of the visual scene and the corresponding sound scene including of the modified first sound object with modified spatial extent.

Abstract: A method comprising: recognizing a first two-handed gesture and a second two-handed gesture in a monitored scene space to at least partially define a shape and position of a computer-implemented virtual boundary in a corresponding monitoring space, wherein the first two-handed gesture in the monitored scene space specifies a first two points in the monitoring space and the second two-handed gesture in the monitored scene space specifies a second two points in the monitoring space; causing implementation of the computer-implemented virtual boundary in the monitoring space corresponding to the monitored scene space, wherein a shape and position of the computer-implemented virtual boundary is at least partially defined by the first two points in the monitoring space and the second two points in the monitoring space; and processing received data to generate a response event when there is a change in a portion of the monitored scene space relative to the computer-implemented virtual boundary in the corresponding m

Abstract: A method for analyzing a presence of objects within a space provided with a capturing system comprising a plurality of camera devices and a playback system for reproducing audio and/or visual signals in the space, the method comprising obtaining a first 3D volumetric representation of a scene within the space, generated on the basis of input streams of at least a first and a second camera device, said first 3D volumetric representation showing at least one object within the scene; sending probe signals to a processing unit; controlling the processing unit to reproduce, using the playback system, one or more audio and/or visual signals on the basis of the probe signals into the space; controlling the processing unit to capture a second 3D volumetric representation of the scene including reproductions of the one or more audio and/or visual signals within the space; and analyzing the reproductions of the one or more audio and/or visual signals captured within the first space whether they correspond to a presumed

Abstract: An apparatus for controlling a controllable position/orientation of at least one audio source within an audio scene, the audio scene including the at least one audio source; a capture device, the apparatus including a processor configured to: receive a physical position/orientation of the at least one audio source relative to a capture device capture orientation; receive an earlier physical position/orientation of the at least one audio source relative to the capture device capture orientation; receive at least one control parameter; and control a controllable position/orientation of the at least one audio source, the controllable position being between the physical position/orientation of the at least one audio source relative to the capture device capture orientation and the earlier physical position/orientation of the at least one audio source relative to the capture device capture orientation and based on the control parameter.

Abstract: A method comprising: using a tracked real point of view of a user in a real space and a first mapping between the real space and a virtual space to determine a point of view of a virtual user within the virtual space; causing rendering to the user at least part of a virtual scene determined by the point of view of the virtual user within the virtual space; and using a selected one of a plurality of different mappings to map tracked user actions in the real space to actions of the virtual user in the virtual space, wherein, when a first mode is selected, the method comprises mapping tracked user actions in the real space, using the first mapping, to spatially-equivalent actions of the virtual user in the virtual space, and wherein, when a second mode is selected, the method comprises mapping tracked user actions in the real space, using a second mapping different to the first mapping, to non-spatially-equivalent actions of the virtual user in the virtual space, wherein the second mapping makes available user i

Abstract: A spatial audio signal is received that is associated with a microphone array configured to provide spatial audio capture and additional audio signal(s) associated with an additional microphone, the additional audio signal having been delayed by a variable delay determined such that common components of the spatial audio signal and the additional audio signal(s) are time aligned. A relative position is received between a first position associated with the microphone array and a second position associated with the additional microphone. Source parameter(s) are received classifying an audio source associated with the common components and/or space parameter(s) identifying an environment within which the audio source is located. Processing effect ruleset is determined based on the source parameter(s) and/or the space parameter(s).