[VCM] Poznań University of Technology Proposal B in response to CfP on Video Coding for Machines
[ 1 ] Instytut Telekomunikacji Multimedialnej, Wydział Informatyki i Telekomunikacji, Politechnika Poznańska | [ P ] pracownik | [ SzD ] doktorant ze Szkoły Doktorskiej | [ S ] student
2022
referat
angielski
EN This is the technical description for Poznań University of Technology Proposal A in response to CfP on Video Coding for Machines. Proposals A and B are virtually the same – they differ by other quantization settings in the VVC encoder. System for Video Coding for Machines where the input (original) and the output (decoded) is a still image or video. The system consists of an encoder, communication/storage system of arbitrary type, and a decoder. The aim of the system is to represent the input image or video with possibly low bitrate but assuring possibly high efficiency of machine vision tasks executed on the output (decoded) image or video. For applications, the machine vision task may comprise a great variety of possible tasks. For example, the goal may be maximization of the average precision of detection or segmentation executed on the output image or video. Another example could be maximization of the tracking performance in the output video. Nevertheless, the applications are not limited to the abovementioned.
1 - 8