Abstract: Systems and methods relating to presenting media content are disclosed. According to an example method, media data comprising a first media content segment is received. The first media content segment comprises first audio data received via the first user device and first video data received via the first user device, and is encoded according to a first set of encoding parameters. A fault is detected in the first media content segment. Detecting the fault comprises determining a quality value of the first media content segment, and determining whether the quality value exceeds a threshold. A second media content segment is received. The second media content segment comprises second audio data and second video data, and is encoded according to a second set of encoding parameters. The first media content segment is replaced with the second media content segment in the media data.

Abstract: A derived quantization parameter for a section of a currently encoding picture of a plurality of pictures is incremented to produce an updated quantization parameter when the derived quantization parameter is less than a minimum quantization parameter for the currently encoding picture. The section is then encoded using the updated quantization parameter. It is emphasized that this abstract is provided to comply with the rules requiring an abstract that will allow a searcher or other reader to quickly ascertain the subject matter of the technical disclosure. It is submitted with the understanding that it will not be used to interpret or limit the scope or meaning of the claims.

Abstract: Various embodiments describe frame selection based on training and using a neural network. In an example, the neural network is a convolutional neural network trained with training pairs. Each training pair includes two training frames from a frame collection. The loss function relies on the estimated quality difference between the two training frames. Further, the definition of the loss function varies based on the actual quality difference between these two frames. In a further example, the neural network is trained by incorporating facial heatmaps generated from the training frames and facial quality scores of faces detected in the training frames. In addition, the training involves using a feature mean that represents an average of the features of the training frames belonging to the same frame collection. Once the neural network is trained, a frame collection is input thereto and a frame is selected based on generated quality scores.

Abstract: Methods, systems, and non-transitory computer readable storage media are disclosed for utilizing deep learning to intelligently determine compression settings for compressing a digital image. For instance, the disclosed system utilizes a neural network to generate predicted perceptual quality values for compression settings on a compression quality scale. The disclosed system fits the predicted compression distortions to a perceptual distortion characteristic curve for interpolating predicted perceptual quality values across the compression settings on the compression quality scale. Additionally, the disclosed system then performs a search over the predicted perceptual quality values for the compression settings along the compression quality scale to select a compression setting based on a perceptual quality threshold. The disclosed system generates a compressed digital image according to compression parameters for the selected compression setting.

Abstract: A method comprising: deriving a first prediction block (608) at least partly based on an output of a neural net (602) using a first set of parameters; deriving a first encoded prediction error block (614-620) through encoding a difference of the first prediction block and a first input block; encoding (620) the first encoded prediction error block into a bitstream; deriving a first reconstructed prediction error block (624) from the first encoded prediction error block; deriving a training signal (628) from one or both of the first encoded prediction error block and/or the first reconstructed prediction error block (624); retraining (630) the neural net (602) with the training signal (628) to obtain a second set of parameters for the neural net (602); deriving a second prediction block (608) at least partly based on an output of the neural net using the second set of parameters; deriving a second encoded prediction error block (614-620) through encoding a difference of the second prediction block and a second

Abstract: Various embodiments describe frame selection based on training and using a neural network. In an example, the neural network is a convolutional neural network trained with training pairs. Each training pair includes two training frames from a frame collection. The loss function relies on the estimated quality difference between the two training frames. Further, the definition of the loss function varies based on the actual quality difference between these two frames. In a further example, the neural network is trained by incorporating facial heatmaps generated from the training frames and facial quality scores of faces detected in the training frames. In addition, the training involves using a feature mean that represents an average of the features of the training frames belonging to the same frame collection. Once the neural network is trained, a frame collection is input thereto and a frame is selected based on generated quality scores.

Abstract: In various embodiments, a perceptual quality application computes an absolute quality score for encoded video content. In operation, the perceptual quality application selects a model based on the spatial resolution of the video content from which the encoded video content is derived. The model associates a set of objective values for a set of objective quality metrics with an absolute quality score. The perceptual quality application determines a set of target objective values for the objective quality metrics based on the encoded video content. Subsequently, the perceptual quality application computes the absolute quality score for the encoded video content based on the selected model and the set of target objective values. Because the absolute quality score is independent of the quality of the video content, the absolute quality score accurately reflects the perceived quality of a wide range of encoded video content when decoded and viewed.

Abstract: A system and method for generating a second reduced size digital image from a first digital image, the method including iteratively compressing the first digital image to an extent determined by a quality measure comprising at least a blockiness measure quantifying added artifactual edges along coding block bom1daries of the second image and/or use of a quantization matrix generated by computing a weighted average of the quantization matrix of the first digital image and a scaled second quantization matrix.

Abstract: An image-processing apparatus and method for variable length coding of header data for image compression, includes retrieval of an input image stored in a first storage space in a memory. The input image is partitioned into a plurality of blocks. A header section is generated comprising a sequence of header bits having a first number of bits for a first block of the plurality of blocks based on application of a sequential encoding scheme on the first block. The header section in which the sequence of header bits may indicate a coding type, a transform type, a prediction type, and a QP value selected in the sequential encoding scheme, is encoded. The selected QP value may be encoded by variable length coding to reduce the first number of bits to a second number of bits in the encoded header section.

Abstract: In one embodiment, a method or system generates a high resolution 3-D point cloud to operate an autonomous driving vehicle (ADV) from a low resolution 3-D point cloud and camera-captured image(s). The system receives a first image captured by a camera for a driving environment. The system receives a second image representing a first depth map of a first point cloud corresponding to the driving environment. The system downsamples the second image by a predetermined scale factor until a resolution of the second image reaches a predetermined threshold. The system generates a second depth map by applying a convolutional neural network (CNN) model to the first image and the downsampled second image, the second depth map having a higher resolution than the first depth map such that the second depth map represents a second point cloud perceiving the driving environment surrounding the ADV.

Abstract: In one embodiment, a method or system generates a high resolution 3-D point cloud to operate an autonomous driving vehicle (ADV) from a low resolution 3-D point cloud and camera-captured image(s). The system receives a first image captured by a camera for a driving environment. The system receives a second image representing a first depth map of a first point cloud corresponding to the driving environment. The system upsamples the second image by a predetermined scale factor to match an image scale of the first image. The system generates a second depth map by applying a convolutional neural network (CNN) model to the first image and the upsampled second image, the second depth map having a higher resolution than the first depth map such that the second depth map represents a second point cloud perceiving the driving environment surrounding the ADV.

Abstract: A system and method for generating a second reduced size digital image from a first digital image, the method including iteratively compressing the first digital image to an extent determined by a quality measure comprising at least a blockiness measure quantifying added artifactual edges along coding block bomldaries of the second image and/or use of a quantization matrix generated by computing a weighted average of the quantization matrix of the first digital image and a scaled second quantization matrix.

Abstract: A display apparatus including: a display; a signal receiver configured to receive a video signal including a plurality of frames; and at least one processor configured to select a main area corresponding to a main color of a first frame among a plurality of divisional areas in the first frame of the video signal, determine whether the first frame is a reference frame based on the size of the selected main area as compared with the first frame, and control the display to display an image based on the video signal in accordance with the determined reference frame.

Abstract: An image determining method for scanning an image and determining specific image pixels of a specific image. The method comprises: determining at least one pixel in a first row having brightness value larger than a threshold value as the specific image pixel; defining a leftmost pixel and a rightmost pixel of the specific image pixel in the first row as a leftmost edge and a rightmost edge of a first specific image range; and defining a second specific image range in a second row of the image, which is next to the first row. Column positions of a leftmost edge and a rightmost edge of the second specific image range are respectively the same with column positions of the leftmost edge and the rightmost edge of the first specific image range. Via this method, the determining for the specific image pixels is more accurate.

Abstract: An image processing apparatus, including processing circuitry configured to decode a bit stream to generate quantized data. The bit stream includes a flag, for each block, that specifies whether or not a difference quantization parameter is present in the bit stream. The flag is included in a first layer that is higher than a second layer in which the difference quantization parameter is set. The processing circuitry is configured to set, according to the flag, a current quantization parameter for a current sub block formed by block partitioning, which splits a block into smaller sub blocks. The processing circuitry is configured to inversely quantize the generated quantized data using the set current quantization parameter.

Abstract: A method of processing an audio signal including at least one audio object based on image information includes: obtaining the audio signal and a current image that corresponds to the audio signal; dividing the current image into at least one block; obtaining motion information of the at least one block; generating index information including information for giving a three-dimensional (3D) effect in at least one direction to the at least one audio object, based on the motion information of the at least one block; and processing the audio object, in order to give the 3D effect in the at least one direction to the audio object, based on the index information.

Abstract: An image processing system, and a method of operation thereof, including: a capture unit for obtaining an input image block; and an image signal processing unit coupled to the capture unit, the image signal processing unit including: a pre-processing module for generating a one-dimensional image block from the input image block, a wavelet transform module, coupled to the pre-processing module, for transforming the one-dimensional image block by wavelet coefficients includes a significant partition and a sub-significant partition identified in the wavelet coefficients, a coding module, coupled to the wavelet transform module, for generating a compressed image block including accessing a variable length coding table based on the significant partition and the sub-significant partition, and an inverse wavelet transform module coupled to the coding module for generating an output image block by performing an inverse wavelet transform on the compressed image block for display on a display device.

Abstract: Methods and systems for compressing images include generating custom quantization tables for quantizing frequency information associated with an image. Specifically, one or more embodiments determine acceptable error percentages during compression of a digital image based on content of the digital image. For example, the acceptable error percentages are defined by compression error thresholds that limit how much error a quantizer in a quantization table can introduce during compression of the digital image. One or more embodiments generate the custom quantization table by determining quantizer values that produce compression errors that meet the compression error thresholds. One or more embodiments compress the digital image using the custom quantization table.

Abstract: A method of DRM protection for an image or a series of images or a succession of video frames, comprises: generating a reversible transform; applying the reversible transform to a source image; compressing the transformed image into a bitstream; and supplying the bitstream to a recipient for consumption. At the recipient the bitstream is decompressed and sent to the image display hardware with the transform still in place. A key is used to generate the inverse transform at the recipient and the inverse transform is then carried out in the image display hardware. An additional transform may be applied to the image following the inverse transform to offset pixels from an average value. Thus the source image itself is never exposed at the recipient although the image can be correctly viewed.

Abstract: An image processing apparatus and a method for processing an image. The image processing apparatus includes circuitry configured to set a current quantization parameter for a current coding block in a layer lower than a layer of a largest coding block and a quantization parameter based on a difference between the current quantization parameter for the current coding block and a slice quantization parameter for a slice in which the current coding block is located. The circuitry quantizes image data to generate quantized data based on the current quantization parameter. Further, the circuitry codes the generated quantized data in a coding block including a hierarchical structure to generate a code data including the quantization parameter.

Abstract: A host device for transmitting print data to a printer is configured to generate bitmap image data by using the print data, to extract object attribute information from objects forming the print data, to convert the generated bitmap image data and the extracted object attribute information into a format of a printer command which is capable of being processed by the printer, and to transmit the converted bitmap image data and the converted object attribute information to the printer.

Abstract: An image processing apparatus and a method for processing an image. The image processing apparatus includes circuitry configured to set a current quantization parameter for a current coding block in a layer lower than a layer of a largest coding block and a quantization parameter based on a difference between the current quantization parameter for the current coding block and a slice quantization parameter for a slice in which the current coding block is located. The circuitry quantizes image data to generate quantized data based on the current quantization parameter. Further, the circuitry codes the generated quantized data in a coding block including a hierarchical structure to generate a code data including the quantization parameter.

Abstract: A system and method for generating a second reduced size digital image from a first digital image, the method including iteratively compressing the first digital image to an extent determined by a quality measure comprising at least a blockiness measure quantifying added artifactual edges along coding block boundaries of the second image and/or use of a quantization matrix generated by computing a weighted average of the quantization matrix of the first digital image and a scaled second quantization matrix.

Abstract: A file transmission method, a file transmission apparatus, and a file transmission system are provided. The method is adapted for a first electronic apparatus to perform a file transmission operation with a second electronic apparatus. The method includes following steps: capturing a target image including a selection region displayed by the second electronic apparatus, wherein the selection region is displayed by the second electronic apparatus in response to a selection operation performed on at least one item displayed by the second electronic apparatus; identifying an encoded pattern located within the selection region in the target image, and decoding the encoded pattern to obtain connection information of the item; and transmitting at least one file with the second electronic apparatus according to the connection information.

Abstract: The present invention provides a method, apparatus and device for image compression. The present invention comprises: determining maximum image reduction information of multiple to be compressed images after compression, based on predetermined compression lower-limit information of the multiple to be compressed images and image size information of each to be compressed image; determining compression parameter information of the multiple to be compressed images based on the maximum image reduction information; determining compression ratio information of each to be compressed image, respectively, based on the compression parameter information, and the image size information and maximum image reduction information after compression of each to be compressed image; compressing each to be compressed image based on the compression ratio information; adding the compressed, multiple to be compressed images into a message to be transmitted for transmission.

Abstract: A video bitstream is received from an encoder, comprising encoded image portions each having a common form representing components of a channel in a color space. Each of a plurality of the encoded image portions comprises a different set of quantized values of the components, including at least a value of a first of the components being quantized from amongst a first scheme of quantized levels. Further, the received bitstream comprises a look-up table mapping the quantized levels of the first scheme to at least partially de-quantized respective levels. A de-quantizer uses the look-up table received in the bitstream to at least partially de-quantize the different quantized values of the first component in a plurality of the image portions, by converting the quantized values of the first component to at least partially de-quantized values corresponding to ones of the at least partially de-quantized levels of the first scheme.

Abstract: An image processing apparatus and a method for processing an image. The image processing apparatus includes circuitry configured to set, based on a value indicating a minimum coding block size for which a difference quantization parameter is set and based on the difference quantization parameter, a current quantization parameter for a current coding block. The current coding block is in a layer that is lower than a layer of a largest coding block. The circuitry is further configured to inversely quantize quantized data based on the set current quantization parameter.

Abstract: Provided are an image reading apparatus, a compression-ratio determination method and a computer-readable, non-transitory medium that can determine a compression ratio such that the data amount after compression would be optimal. The image reading apparatus includes an image generator for generating an input image by reading a document, a document-type identification module for identifying a type of the document, and an image compressor for compressing the input image, wherein the image compressor determines a compression ratio for compressing the input image, based on the type of the document.

Abstract: A video transmission system includes a transceiver module that transmits a video signal to a remote device over at least one communications channel wherein the video signal is transmitted as at least one separate video layer stream chosen from, an independent video layer stream and at least one dependent video layer streams that require the independent video layer for decoding. A control module determines at least one channel characteristic of the at least one channel and chooses the at least one separate video layer stream based on the at least one channel characteristic of the at least one channel.

Abstract: A size comparison unit reports, to an access control unit, the size of a piece of compressed data having the smallest size from among pieces of compressed data generated via compression processing units compressing blocks, an algorithm name representing a compression processing unit that has generated the piece of compressed data having the smallest size; the access control unit selects a piece of compressed data to be written in a tape medium and reports, to a statistical information processing unit, an algorithm name representing a compression processing unit that has generated this selected piece of compressed data; the statistical information processing unit extracts, from the reported algorithm name, a regularity present in a result of the selection of compressed data and, in accordance with this extracted regularity, specifies and stops a compression processing unit to be stopped.

Abstract: This invention is directed to reduce the correction load on a user when a plurality of images are to be grouped. An information processing method of discriminating an object contained in a first image and grouping the first image to a plurality of grouping destinations for each object includes a discrimination step of discriminating an object from the first image, a grouping destination determination step of determining, on the basis of the discrimination result obtained in the discrimination step, to which one of the plurality of grouping destinations the first image is to be grouped, and a display control step of displaying, on a window, a process of moving the first image to a position of the second image corresponding to the grouping destination determined in the grouping destination determination step.

Abstract: Methods for managing an interactive streaming image system are disclosed. More particularly, hardware and/or software for generating, encoding, and transmitting image frames to an interactive client are disclosed. One embodiment provides a method for streaming images from a server to a client. Embodiments may include receiving from the client via a network client information and generating a new image frame based on the received client information. Embodiments may also include comparing the new image frame and a previous image frame and setting a new compression level based on the comparison between the new image frame and the previous image frame. Embodiments may also include encoding the new image frame based on the new compression level and transmitting the encoded new image frame and an indication of the new compression level to the client via the network.