Artificial neural networks have been adopted for a broad range of tasks in multimedia analysis and processing, such as visual and acoustic classification, extraction of multimedia descriptors or image and video coding. The trained neural networks for these applications contain a large number of parameters (i.e., weights), resulting in a considerable size. Thus, transferring them […]
Neural Network Compression Read More »
In July 2021, MPEG has completed development of the MPEG Immersive Video (MIV) standard. MIV was developed to support compression of immersive video content in which multiple real or virtual cameras capture a real or virtual 3D scene. The standard enables storage and distribution of immersive video content over existing and future networks for playback
MPEG Immersive Video Read More »
MPEG-5 will be a video coding standard to address business and technology needs in use cases that are not well served by existing ISO standards, such as HEVC. The goal is to develop a new video coding standard that meets a combination of business and technology requirements: In particular Part 2 of ISO/IEC 23094 which
MPEG-5 Video (ISO/IEC 23094) Read More »
MPEG announced the completion of the new Versatile Video Coding (VVC) standard at its 131st meeting in July 2020. VVC is the latest in a series of very successful standards for video coding that have been jointly developed with ITU-T, and it is the direct successor to the well-known and widely used HEVC (Rec. ITU-T H.265
Versatile Video Coding Read More »
MPEG-I Audio seeks to provide an immersive virtual audio and visual experience to users wearing a head-mounted visual display (HMD) with audio headphones, or using loudspeakers. The user’s view and perception of the virtual reality environment (VR) is responsive to user motion (i.e. position and orientation), where the user is free to move around in
Immersive Audio with 6 Degrees of Freedom Read More »
The phrase “Internet of Things” (IoT) encompasses a large variety of research, development and market efforts related to the communication between smart objects. The definition may be fuzzy, but the market reality is very clear: the number of devices connected to the Internet will reach 50 billion by 2020. An important factor contributing to the
Internet of Media Things (IoMT) Read More »
In October 2020, MPEG completed development of its ISO/IEC 23090-9 Geometry-based Point Cloud Compression (G‑PCC) standard. G‑PCC addresses lossless and lossy coding of time-varying 3D point clouds with associated attributes such as colour and material properties. This technology is particularly suitable for sparse point clouds. ISO/IEC 23090-5 Video-based Point Cloud Compression (V‑PCC), which was completed
Geometry-based Point Cloud Compression Read More »
MPEG has issued a Call for Proposals (CfP) on Dynamic Mesh Coding, and expects to evaluate responses to this call in April 2022. A dynamic mesh sequence may require a large amount of data since it may consist of a significant amount of information changing in time. Therefore, efficient coding technologies are required to store
Dynamic Mesh Coding Read More »
ISO/IEC 23001-11 specifies metadata (Green Metadata) that facilitates the reduction of energy usage during media consumption. The metadata enable means for (a) reducing decoder power consumption, (b) reducing display power consumption, (c) media selection for the joint decoder and display power reduction, and (d) quality recovery after low-power encoding. The third edition, which is currently
Energy-Efficient Media Consumption (Green Metadata) Read More »
In July 2020, MPEG completed its development of a Video-based Point Cloud Compression (V-PCC) standard. V-PCC addresses lossless and lossy coding of 3D point clouds with associated attributes such as colors and reflectance. Point clouds are typically represented by extremely large amounts of data, which is a significant barrier for mass market applications. However, the
Video-based Point Cloud Compression Read More »
