SYSTEM AND METHOD FOR IMAGE ENCODING

    公开(公告)号:US20250133215A1

    公开(公告)日:2025-04-24

    申请号:US18901464

    申请日:2024-09-30

    Applicant: Axis AB

    Abstract: A method of encoding images in a video, comprises: acquiring an original image from an image sensor of a video camera; encoding the original image using a generative image model, thereby obtaining a first encoded image; decoding the image to obtain a first decoded image; identify a region of interest (ROI) of the original image; for each ROI: perform an encoding quality check by comparing several reference points in the ROI of the original image against corresponding reference points in the ROI of the first decoded image, thereby obtaining a difference; if the difference is greater than a threshold, encoding the ROI using a non-generative image model, thereby obtaining a non-generative encoded image area; providing final encoded image data comprising a) the non-generative encoded image areas for the ROI having a difference greater than the threshold and b) the first encoded image for a remaining part of the original image.

    Method, a computer program, an encoder and a monitoring device

    公开(公告)号:US11545160B2

    公开(公告)日:2023-01-03

    申请号:US16863439

    申请日:2020-04-30

    Applicant: Axis AB

    Abstract: A method, a computer program product, an encoder and a monitoring device for encoding an audio signal with variable bitrate, wherein: an audio signal comprising a plurality of successive audio frames is received; and for each successive audio frame of the audio signal: the audio frame is represented in a frequency domain with respect to a plurality of frequency sub-bands; the audio frame is classified in each frequency sub-band as either background or foreground using a background model specific to the frequency sub-band; each successive audio frame of the audio signal is encoded, wherein a number of bits is allocated for each frequency sub-band of the audio frame, wherein the number of bits allocated for a frequency sub-band is higher if the audio frame is classified as foreground in the frequency sub-band than if the audio frame is classified as background in the frequency sub-band.

    Method and bitrate controller for controlling output bitrate of a video encoder

    公开(公告)号:US11178401B2

    公开(公告)日:2021-11-16

    申请号:US16863245

    申请日:2020-04-30

    Applicant: Axis AB

    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.

    Using a skip block mask to reduce bitrate from a monitoring camera

    公开(公告)号:US11055976B2

    公开(公告)日:2021-07-06

    申请号:US16947659

    申请日:2020-08-11

    Applicant: Axis AB

    Abstract: Methods and apparatus, including computer program products, implementing and using techniques for reducing bitrate from a monitoring camera. A first input is received that identifies first regions of an image representing a camera field of view. The first regions contribute significantly to the bitrate. A second input is received that identifies second regions of the image. The second regions contain information that is deemed to be of little visual interest to a user of the monitoring camera. Third regions of the image are determined. The third regions are regions where the first and second regions overlap at least in part. Video encoder settings are applied to force skip blocks in at least some of the third regions, thereby reducing contributions to the bitrate from the third regions.

    METHOD AND ENCODER SYSTEM FOR DETERMINING GOP LENGTH FOR ENCODING VIDEO

    公开(公告)号:US20180270482A1

    公开(公告)日:2018-09-20

    申请号:US15921302

    申请日:2018-03-14

    Applicant: Axis AB

    Abstract: A method of determining a maximum permissible key frame distance for use in encoding image frames in a video sequence comprising encoding a plurality of video test sequences, each video test sequence having a respective noise level, where each video test sequence is encoded using a respective quantization parameter and a plurality of key frame distances is disclosed. For each video test sequence, an output bitrate is determined for each key frame distance. For each combination of noise level, quantization parameter, and key frame distance, the determined output bitrate is stored and a maximum key frame distance at which output bitrate decreases in response to an increase in key frame distance is identified as maximum permissible key frame distance. A method of encoding image frames in a video sequence is also disclosed, as well as a system, a camera and a computer program product for performing the methods.

Patent Agency Ranking