Getting the Word Out On MPEG7/VCM for Vision Based Applications using AI (Edge & Cloud)

Computer vision technology is poised for incredible growth across nearly all applications. Machine-to-Machine (M2M) connections and human-machine (H2M) vision will represent the greatest usage of Internet video traffic. Visual intelligence is being incorporated into new products and practical applications, including automotive, smart cites, security and surveillance, consumer electronics, industrial, health, retail and beyond.

Early this year, MPEG Chairman Leonardo Chiariglione announced the formation of a next-gen video codec standard: “MPEG Video Coding for Machines (VCM)”. MPEG-VCM technology, as an intelligent video codec standard for M2M and H2M vision, is broadly predicted to be the dominant source and use of IP traffic, particularly in the emerging era of 5G massive M2M (IoT), low latency and broadband communications.

Traditional video coding methods aim for the best video/image quality under certain bit-rate constraints and tailored for human consumption. However, with the rise of machine learning applications, many vision-based intelligent capabilities are being developed into new and existing systems and applications. Any use cases in which video features need to be transmitted for additional processing and that may potentially be used for machine or human end users could benefit from a standard in the coded features (i.e., a shared backbone).

The MPEG-VCM standard aims for the compressed coding of this feature stream for efficient transmissions (low latency) and archiving concerns (storage). Currently, the emerging MPEG-VCM standard is focusing on the intelligent transportation and connected vehicles, video surveillance, and smart city as its use cases.

Industries need a standard to address and support adaptation with interoperable solutions which benefits businesses, governments and end-users. Over 200 industry leaders and experts have already expressed their interests by subscribing to MPEG-VCM (https://lists.aau.at/mailman/listinfo/mpeg-vcm).

Interested Embedded Vision Alliance companies and other community members are invited to participate in driving innovations in the next-gen AI-based video codec standard. Companies making this commitment will play a significant role in shaping the future of AI solutions across industries.

The upcoming MPEG-VCM (MPEG129) is scheduled for Jan. 12-17, 2020 in Brussels

MPEG 129 – Brussels: https://mpeg.chiariglione.org/meetings/129.

Sign Up for Access to VCM Archives: https://lists.aau.at/mailman/listinfo/mpeg-vcm

Marc Naddell
Vice President of Marketing, Gyrfalcon Technology

Here you’ll find a wealth of practical technical insights and expert advice to help you bring AI and visual intelligence into your products without flying blind.

Contact

Address

Berkeley Design Technology, Inc.
PO Box #4446
Walnut Creek, CA 94596

Phone
Phone: +1 (925) 954-1411
Scroll to Top