10 results about "Variable bit rate vbr" patented technology

The invention discloses a dynamic self-adaptive edge-cloud cooperative computing segmentation point and compression strategy adjustment method. The core of the method is to deploy a dynamic decision engine on edge equipment. The engine senses a network state (such as bandwidth and delay) and a task priority specified by an application in real time, and according to a preset decision logic (such as a look-up table or a lightweight prediction model), a deep learning model which supports multiple segmentation points and is equipped with a variable bit rate hierarchical quantization feature compression model for each segmentation point is used for performing a multi-segmentation-point hierarchical quantization feature compression model on the basis of the multi-segmentation-point hierarchical quantization feature compression model, so that the multi-segmentation-point hierarchical quantization feature compression model is obtained. And dynamically selecting an optimal model segmentation point and feature compression level combination. The combination is transmitted to the cloud, and the cloud loads the corresponding back-end model to complete calculation. Through the dynamic adjustment mechanism, the real-time optimal balance among the reasoning delay, the task precision and the network overhead is realized, and the performance stability, the resource utilization rate and the adaptability of the system in a variable environment are remarkably improved.

The invention discloses a token stream guide variable code rate image compression method oriented to unification of perception and understanding. The method comprises the following steps of: 1, acquiring a two-dimensional image, and processing the two-dimensional image to acquire a one-dimensional token sequence; 2, processing the one-dimensional token sequence through a variable token mask to generate a binary compressed bit stream; 3, decoding the compressed bit stream to obtain an unmasked token sequence, and carrying out dynamic prediction on the unmasked token sequence to recover a complete token sequence; and 4, realizing human perception through the receiving end I after the complete token sequence is recovered, realizing machine perception through the receiving end II after the complete token sequence is recovered, and finally realizing image compression. The method has the capability that a single model supports the continuous variable bit rate, supports a large language model to directly perform semantic understanding based on the compressed code stream, and has the characteristics of low calculation complexity and low delay.

The application provides a lightweight variable bit rate multi-view image compression method and system, and relates to the technical field of image compression. The method comprises: downsampling, feature extraction and feature scaling of a single-view image to obtain a latent representation corresponding to a target bit rate; quantization and lossless entropy coding of the latent representation to obtain a final compressed bit stream; subsequent lossless entropy decoding and inverse scaling to restore the latent representation; feature fusion and upsampling of the restored latent representations of different views to generate a reconstructed compressed image. The model comprises a main encoder, a feature scaling module, a quantization module, an autoregressive entropy model, an arithmetic encoder, an arithmetic decoder, a feature inverse scaling module and a decoder. The application can efficiently compress image data, reduce computational complexity and storage space occupation while preserving image details and quality, thereby providing faster speed and lower bandwidth requirement for image transmission and storage.

An autoencoder and a method for adaptive learned image compression with configurable encoder and decoder for variable bitrate applications are described, wherein the encoder comprises: - a pre-trained transformer-based architecture operable to convert input image data into quantized latent representations (9); - one or more low-rank adapter modules (W a,W b) integrated within said encoder (g a), each adapter module (W a,W b) being configured to adjust the encoding process for different target bitrates while maintaining pre-trained model parameters; - wherein said one or more low-rank adapter modules (W a,W b) are incorporated into fully connected layers (120,121) of the encoder's multi-layer perception modules (MLP), enabling said encoder (g a) to achieve efficient rate-distortion performance across a range of bitrates by fine-tuning only parameters of said adapter modules (W a,W b); - a mechanism adapted to merge said adapter modules (W a,W b) with pre-trained model weights following adaptation, thereby restoring the encoder's complexity to its original level.

Image capture devices and methods may be used to pre-buffer media storage. The pre-buffering method includes recording an image capture segment of a variable bitrate input stream in a circular buffer. The circular buffer includes a number of recordable segments. The method includes recording a next image capture segment in a next adjacent recordable segment of the circular buffer if the next adjacent recordable segment of the predetermined number of recordable segments is available. The method includes overwriting an oldest recordable segment if the next adjacent recordable segment of the predetermined number of recordable segments is not available.

The application discloses a dynamic self-adaptive edge-cloud collaborative computing segmentation point and compression strategy adjustment method. The core of the method is to deploy a dynamic decision engine on the edge device. The engine real-time perceives network state (such as bandwidth, delay) and application specified task priority, and according to the preset decision logic (such as query table or lightweight prediction model), dynamically selects the optimal model segmentation point and feature compression level combination from a deep learning model supporting multiple segmentation points and equipped with variable bit rate layered quantization feature compression model for each segmentation point. The combination is transmitted to the cloud, and the corresponding backend model is loaded to complete the calculation. Through the dynamic adjustment mechanism, the application realizes the real-time optimal balance among inference delay, task accuracy and network overhead, and significantly improves the performance stability, resource utilization and adaptability of the system in a variable environment.

Various systems and methods for providing variable bitrate compression for split deep neural network (DNN) computing are described herein. A system may be configured to manage a split DNN, the split DNN configured to operate on a compute system and a second system over a communication network. The system may access a performance metric; determine, based on the performance metric, a split point of the split DNN, the split point defining a head portion of the split DNN and a tail portion of the split DNN; determine, based on the performance metric, a bottleneck layer configuration for a bottleneck layer at the split point, the bottleneck layer including a bottleneck encoder and a bottleneck decoder; execute the head portion of the DNN and the bottleneck encoder on the compute system; and recurrently access an updated performance metric and determine a revised split point or a revised bottleneck layer configuration based on the updated performance metric.

Provided are methods and systems for providing variable bitrate content (e.g., video content, audio content, multimedia content, etc.). The content may comprise a plurality of portions (e.g., frames, segments, fragments, etc.). Each portion of the content may be tagged and / or associated with a content element. A content element may be associated with and / or indicate one or more of an encoding parameter associated with a respective portion of the content, attributes of the content (e.g., a scene transition, scene change, etc.) associated with the respective portion of the content, or additional content related items (e.g., one or more advertisements, etc.) associated with the respective portion of the content. The content element may be used to determine a bitrate to associate with the respective portion of the content. The content may be received / retrieved according to the determined bitrates.

Systems, methods, and instrumentalities are disclosed herein for quantization processes used within image or video compression that perform end-to-end learning with additional parameters (e.g., quantization offsets). A video decoding and / or encoding device may include a processor. The device may be configured to obtain latent variables associated with a neural network (NN) model. The device may obtain a quantization offset associated with the latent variable. The device may reconstruct a quantized latent variable associated with the NN model based on the quantization offset. The quantization offset associated with the latent variable may be obtained based on at least one of a lookup table, using a second neural network, or an estimated standard deviation based on a distribution associated with the latent variable.

A method, system, and computer program to encode and decode a channel coherence parameter applied on a frequency band basis, where the coherence parameters of each frequency band form a coherence vector. The coherence vector is encoded and decoded using a predictive scheme followed by a variable bit rate entropy coding.