Back to Search Document details
23rd Meeting: by teleconference, July 2021 2021-07-01 03:06
EE1-2.3: Neural Network-based Super Resolution
Authors: Anand Meher Kotra Qualcomm Kevin Reuzé Qualcomm Jie Chen Qualcomm H. Wang Qualcomm Marta Karczewicz Qualcomm Jingxin Li Qualcomm Hongtao Wang
Abstract
This contribution reports the EE test results of JVET-V0096. JVET-V0096 studied the performance of applying a Neural-Network based super-resolution used as upsampling filter in the context of VVC RPR. Prior to encoding, a given picture is downsampled by a factor of 2x using the inbuilt RPR mechanism of VTM11. PSNR of the coded frame is computed by calculating the MSE between the original picture and the upsampled version of the decoded picture. The upsampled picture is generated by the Neural Network-based upsampling filter instead of the existing VTM upsampling filter.
Meeting Notes
JVET-W0105 EE1-2.3: Neural Network-based Super Resolution [A. M. Kotra, K. Reuzé, J. Chen, H. Wang, M. Karczewicz, J. Li (Qualcomm)]
References:
JVET-T0092 JVET-U0099 JVET-V0096 JVET-V2016 JVET-V2023
Cited:
JVET-W0011 JVET-W0023 JVET-W0182 JVET-W0186 JVET-W2023 JVET-X0107 JVET-X2023
PATENTS:
20250259266 0.72 WO/2025/171030 0.34 20250324100 0.33 WO/2025/216864 0.32 20250285222 0.30 WO/2024/159502 0.28 WO/2024/159480 0.27 4064133 0.26
ARCHIVE v1 :
JVET-W0105-v1.docx JVET-W0105-v1.xlsm
Decisions
JVET-W0105 EE1-2.3: Neural Network-based Super Resolution [A. M. Kotra, K. Reuzé, J. Chen, H. Wang, M. Karczewicz, J. Li (Qualcomm)]
Citation