Back to Search Document details
39th Meeting: Daejeon, KR, June 2025 2025-07-03 02:18
Suggested text and figure on Call for Evidence document
Abstract
As a manufacturing company, decoder complexity is very important as well as encoder complexity. In the JVET-AM0237, it is suggested to illustrate both encoding and decoding times to provide a comprehensive understanding of trade-offs within the constrained encoding category. Having decoding time information discussed jointly with encoding time, it is beneficial for CfE response and practical implementation of the next generation of video coding standard.
JVET-AM0338 Suggested text and figure on Call for Evidence document [L. Li, W. Choi, M. Budagavi, K. P. Choi (Samsung)] [late]

As a manufacturing company, decoder complexity is very important as well as encoder complexity. In the JVET-AM0237, it is suggested to illustrate both encoding and decoding times to provide a comprehensive understanding of trade-offs within the constrained encoding category. Having decoding time information discussed jointly with encoding time, it is beneficial for CfE response and practical implementation of the next generation of video coding standard.

This contribution suggests text and figure on decoding time for providing an example of an decoder run time vs compression performance curve in CfE document.It was commented that without doubt that decoding runtime is an important factor. However, the main aspect of that test is proof of existence of runtime-constrained encoding, and decoder speed may be quite independent from that, and it can hardly be expected that it varies as much as the encoding time as shown in the contribution.

Decoder runtime needs to be reported anyway along with the encoder run time, but JVET should retain the freedom to decide later how those data will be evaluated and be illustrated in the CfE report. Therefore, it was agreed to have no example figure about a hypothetical method of assessment in the call.

AHG18 Ultra-low latency and packet loss resilience (12)

Contributions in this area were discussed during 1900–2030 on Monday 30 June 2025 and during 1415–1450 on Tuesday 1 July (chaired by JRO).

Decisions
Decoder runtime needs to be reported anyway along with the encoder run time, but JVET should retain the freedom to decide later how those data will be evaluated and be illustrated in the CfE report. Therefore, it was agreed to have no example figure about a hypothetical method of assessment in the call.
Citation