Abstract: A method of controlling output bitrate of a video encoder encoding a video sequence comprises: setting an allowable average bitrate, gathering encoding data by encoding video during a first time period and for each of a plurality of time intervals, storing a respective output bitrate, and a quantisation parameter used for encoding during the respective time interval. For each time interval, the output bitrate is normalised by calculating a corresponding normalised bitrate estimated for a predetermined nominal quantisation parameter as a function of the output bitrate and the quantisation parameter. An average normalised bitrate is calculated by averaging the normalised bitrates of the plurality of time intervals. A new quantisation parameter is calculated based on a comparison of the average normalised bitrate and the allowable average bitrate. The video sequence is encoded using the new quantisation parameter during a second time period subsequent to the first time period.

Abstract: A video encoding method for encoding a series of images of original video data into an encoded video, comprises setting a periodic intra refresh pattern for the encoding of the series of images; setting a number of delta frames to be added between consecutive intra refresh frames of the periodic intra refresh pattern, thereby adjusting the intra refresh periodicity of the encoded video stream; and encoding the series of images of original video data into the encoded video stream using the set periodic intra refresh pattern and the set number of delta frames to be added between consecutive intra refresh frames. During the encoding, a current intra refresh frame is set to directly refer to a previous intra refresh frame and each delta frame added in between two consecutive intra refresh frames is set to only refer to a previous one of the two consecutive intra refresh frames.

Abstract: A method for encoding, using a block-based video encoding algorithm, a distorted image frame produced via at least one image sensor, comprising: determining a map of maximum pixel block sizes corresponding to the distorted image frame, wherein the map of maximum pixel block sizes is determined based on a spatial resolution distribution corresponding to a distribution of extents of the corresponding FOV, such that for a first portion of the distorted image frame having a first spatial resolution, the maximum pixel block size corresponding to the first portion is set to a first value, and for a second portion of the distorted image frame having a second spatial resolution being lower than the first spatial resolution, the maximum pixel block size corresponding to the second portion is set to a second value being lower than the first value; and encoding, using the block-based video encoding algorithm, the distorted image frame.

Abstract: A method and an apparatus for forming a video stream based on images of a scene captured by a camera. The method comprises: encoding image frames based on images captured by the camera, forming an encoded portion; determining a length of motion for a sample image frame, the length of motion being indicative of the amount of relative motion between the camera and the scene in the sample image frame relative a temporally preceding image frame; setting a motion length threshold based on a motion vector search range of the encoder; comparing the determined length of motion with the set motion length threshold; based on the outcome of the comparison, adjusting a chosen temporal frame distance of at least an adjustment portion (PA) of the video stream, wherein the adjustment portion is comprised in the encoded portion or in a temporally later encoded portion.

Abstract: Methods and apparatus, including computer program products, implementing and using techniques providing privacy masks on a video stream. In a video received from a monitoring camera, at least one image region is selected, to which region a privacy mask should be applied. For any image frame in the video that contains the at least one image region, a flag is set to indicate a non-displayed image frame. A forward-predicted image frame referencing the non-displayed image frame is inserted, wherein coding units representing the at least one image region in the forward-predicted image frame collectively provide the privacy mask by obfuscating the at least one image region.

Abstract: A method for detecting events comprises repeatedly registering a value indicating an amount of data generated by an encoder, which is encoding video from a scene by means of temporal compression, determining if a particular event has occurred in the scene represented by the encoded video by comparing characteristics of the registered values with predetermined characteristics, and generating an event signal in response to an event occurrence being determined.

Abstract: A method, system, and non-transitory computer readable medium for encoding a privacy masked image into an encoded image frame of a video stream is disclosed. The encoding comprises receiving pixel divided image data of an image; receiving data defining an area and graphical characteristic of a privacy mask that masks part of the image; extending the privacy mask to be aligned with one or more encoding units having one or more pixels located within the privacy mask to form a redefined privacy mask; generating a privacy masked image by applying the redefined privacy mask and the graphical characteristic of the privacy mask to the image; and encoding the privacy masked image into the encoded image frame of the video stream, wherein the encoding comprises preventing spatial prediction encoding and/or temporal prediction encoding of pixels within the redefined privacy mask of the image in response to certain conditions.

Abstract: A method encodes an image of a video sequence by analyzing motion in the video sequence to identify a coherent region in the image that has moved with a distance vector which is longer than a threshold since a previously encoded image in the video sequence. A cost for inter-coding the block is calculated as a combination of a cost for encoding a motion vector and a residual. For at least one block of pixels in the coherent region, the cost for encoding the motion vector is calculated relative to the distance vector of the coherent region instead of being calculated relative to a motion vector of a neighbouring block of pixels in the image. The block is then inter-coded if the combined cost for encoding the motion vector and the cost for encoding the residual when combined are below a cost for intra-coding the block of pixels.

Abstract: A method for encoding and transmitting a plurality of image frames of a video stream includes: intra refresh encoding the plurality of image frames; encoding one image frame of the plurality of image frames to produce encoded image data of the one image frame, the encoded image data being organized in a plurality of data packets, wherein the encoding comprises intra frame encoding the one image frame; transmitting, over a time period, the intra refresh encoded image frames; transmitting the plurality of data packets, wherein the transmission of the plurality of data packets is distributed over the time period during which the intra refresh encoded image frames are transmitted. An interrelated method for receiving and storing the plurality of image frames, as well as devices and computer program products for performing the methods, are also provided.

Abstract: Video encoding is provided using periodic intra refresh, which can adapt padding of an intra encoding region between image frames based on a current level motion in the image frame. Upon a current level of motion not exceeding a first threshold, an intra encoding region of the image frame according to a predefined periodic intra refresh pattern is intra encoded. Upon the current level of motion exceeding the first threshold, the intra encoding region of the image frame is expanded to overlap to a determined extent with a previously encoded intra encoding region among the plurality of image frames, and the expanded intra encoding region of the image frame is intra encoded. Regions of the image different from the intra encoding regions and the expanded intra encoded region, respectively are inter frame encoded.

Abstract: The present invention relates to a method for adding a graphical object to image data of one or more image frames comprising a first image frame, the graphical object being added during encoding of the one or more image frames using an encoder, wherein the image data generator is converting the graphical object into a first set of discrete frequency transform, DFT, blocks each DFT block being associated with a position in the one or more image frames to which the DFT block corresponds. The encoder is performing DFT on a plurality of block of pixels of the first image frame, to output a second set of DFT blocks and then replacing a subset of DFT blocks of the second set of DFT blocks with the first set of DFT blocks according to the positions associated with the first set of DFT blocks.

Abstract: A video encoding method for encoding a series of images of original video data into an encoded video, comprises setting a periodic intra refresh pattern for the encoding of the series of images; setting a number of delta frames to be added between consecutive intra refresh frames of the periodic intra refresh pattern, thereby adjusting the intra refresh periodicity of the encoded video stream; and encoding the series of images of original video data into the encoded video stream using the set periodic intra refresh pattern and the set number of delta frames to be added between consecutive intra refresh frames. During the encoding, a current intra refresh frame is set to directly refer to a previous intra refresh frame and each delta frame added in between two consecutive intra refresh frames is set to only refer to a previous one of the two consecutive intra refresh frames.

Abstract: A method for encoding video is disclosed wherein information representative of pixels in an image frame is received, and a spatial statistical measure of said information is calculated for groups of neighbouring pixels to form a group value for each group of pixels. A set of available quantization steps is determined comprising a first predetermined quantization step. For a first group of neighbouring pixels, the method comprises: for each available quantization step calculating a remainder after division of the group value of the first group of pixels by the respective available quantization step. The quantization step of the set that results in the smallest remainder is selected as selected quantization step. The first group of pixels is encoded using the selected quantization step. A method of encoding differences between image frames is also disclosed, as well as encoding systems.

Abstract: A method encodes a sequence of image frames depicting an object whose privacy is to be protected. A first and a second version of the sequence of image frames is generated. Image frames of the first version, but not the second version, of the sequence are provided with a privacy mask to protect the privacy of the object depicted therein. The first version of the sequence is encoded as a first encoded sequence of intra-coded image frames and inter-coded image frames. A second version of the sequence is encoded as a second encoded sequence of inter-coded image frames. The inter-coded image frames of the second encoded sequence are encoded by using the intra-coded image frames of the first encoded sequence. The inter-coded image frames of the second encoded sequence are encrypted, and thereafter the first and the second encoded sequence are merged into a single output sequence.

Abstract: Methods and apparatus, including computer program products, implementing and using techniques providing privacy masks on a video stream. In a video received from a monitoring camera, at least one image region is selected, to which region a privacy mask should be applied. For any image frame in the video that contains the at least one image region, a flag is set to indicate a non-displayed image frame. A forward-predicted image frame referencing the non-displayed image frame is inserted, wherein coding units representing the at least one image region in the forward-predicted image frame collectively provide the privacy mask by obfuscating the at least one image region.

Abstract: A video encoding method comprises encoding a series of images of original video data into an encoded video stream comprising key frames and delta frames, wherein the delta frames are organized in a hierarchical prediction pattern comprising a plurality of temporal layers. The video encoding method further comprises adding to the encoded video stream a hidden delta frame for at least some of the key frames. Each hidden delta frame corresponds to a key frame, is based on same original video data as the corresponding key frame and is referring to a previous key frame in the encoded video stream. Also, a method of reducing a file size of the video stream encoded according to the video encoding method is disclosed.

Abstract: Methods and apparatus, including computer program products, implementing and using techniques for encoding a video sequence comprising a plurality of image frames, by an encoder are described. An image frame is received from a video stream. An input is received, which indicates one or more regions in the received image frame for which a privacy mask should be applied. The one or more regions are represented by one or more coding units. The image frame is encoded into an output frame, wherein image data in the one or more regions is replaced by intra-predicted coding units with transformed coefficients set to zero, the intra-predicted coding units are obtained from a prediction stage in the encoder.

Abstract: A technique for encoding a privacy masked image comprising a plurality of pixels into an encoded image frame of a video stream is disclosed, The technique includes receiving image data of the image, wherein the image data represents the pixel of the image, grouping the pixels of the image into encoding units, wherein each encoding unit comprises a plurality of neighboring pixels; receiving data defining a privacy mask masking part of the image in the interest of personal integrity, and graphical characteristics of the privacy mask including any one or a combination of constant coloring, smearing, and pixelization, identifying all encoding units having one or more pixels located within the privacy mask, extending the privacy mask to be aligned with the identified encoding units to form a redefined privacy mask covering the identified encoding units, generating a privacy masked image by applying the redefined privacy mask and the received graphical characteristics to the image, and encoding the privacy masked i

Abstract: The present teachings relate to the field of video encoding. In particular, the present teachings relate to a method, device, and system for encoding a sequence of frames wherein a motion level of image data of an initial frame in the sequence of frames is determined to be below a motion level threshold. The encoding method described herein may reduce the bit rate for the encoded sequence of frames, while avoiding that the encoded video stream looks unnatural or frozen when decoded and displayed.

Abstract: A method of controlling output bitrate of a video encoder encoding a video sequence comprises: setting an allowable average bitrate, gathering encoding data by encoding video during a first time period and for each of a plurality of time intervals, storing a respective output bitratet, and a quantisation parameter used for encoding during the respective time interval. For each time interval, the output bitrate is normalised by calculating a corresponding normalised bitrate estimated for a predetermined nominal quantisation parameter as a function of the output bitrate and the quantisation parameter. An average normalised bitrate is calculated by averaging the normalised bitrates of the plurality of time intervals. A new quantisation parameter is calculated based on a comparison of the average normalised bitrate and the allowable average bitrate. The video sequence is encoded using the new quantisation parameter during a second time period subsequent to the first time period.