Hey Teams
We thank all of you for putting your effort and time into our challenge.
The results of the final subjective evaluation for the NTIRE 2025 Challenge on Video Quality Enhancement for Video Conferencing are now available. You can checkout the scores on the CodaLab webpage [https://codalab.lisn.upsaclay.fr/competitions/21291#:~:text=at%20CVPR%202025.-,Subjective%20Results,-We%20received%205].
You can find more detailed, pairwise statistics here [https://github.com/varunj/cvpr-vqe/tree/main/results]. There were some submissions with similar names, so I modified some names. The mapping can be found in our spreadsheet [https://docs.google.com/spreadsheets/d/1r4xdrKz7pKM_NnDdHv7AYuIh3EqR-TAPivXNPbpPTS0].
We thank the top teams for submitting their fact sheets. We are now working on compiling the challenge report. Please let me know if you see any discrepancies in the results.
Finally, we invite all the teams to consider the NTIRE 2025 workshop as a venue for publishing original methods presented in this challenge. The paper submission deadline for challenge papers is Mar 28, 2025 [https://cvlai.net/ntire/2025/]. Each paper will be reviewed following the standard CVPR review process. When referring to the challenge in your paper, we encourage the authors to use the following BibTeX format:
@inproceedings{jain2025ntire,
title={NTIRE 2025 Challenge on Video Quality Enhancement for Video Conferencing: Datasets, Methods and Results},
author={Jain, Varun and Wu, Zongwei and Zou, Quan and Florentin, Louis and Turbell, Henrik and Siddhartha, Sandeep and Timofte, Radu and others},
booktitle={Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition},
year={2025}
}
We hope to meet all of you in Nashville during the NTIRE 2025 workshop, at CVPR 2025.
Best
The Challenge Organizers