Download the test set
- The test set includes 80 video clips, each with 300 frames for each track.
- Link to the list of clips in the test set will be sent to registered teams (hereafter trackX_testset.txt).
- Use the utility (opens in new tab) script (opens in new tab) (opens in new tab) from the challenge’s Git repository and thetrackX_testset.txt file to download the test set.
- Example command:
python downloader.py --list-of-files trackX_testset.txt --local-path LOCAL_PATHFile naming and format
Each track consists of 80 test clips, and for each clip, we require an upscaled and enhanced version in return. The processed clip should retain the same name as the input clip. Ensure the output clips are encoded using H.264 with a CRF = 10, a pixel format of YUV420p, and 30FPS (similar to the input).
Example FFMPEG command
ffmpeg -framerate 30 -i [PNGG_DIR]/%d.png -y -c:v libx264 -crf 10 -preset veryslow -pix_fmt yuv420p output.mp4Paper submission
Participants in the grand challenge must submit a paper following the same format as the main ICME conference (see details here (opens in new tab)) but as single-blind submissions. Papers must be no longer than 6 pages, including all text, figures, and references. The papers should describe their models, any additional data used in training, and the complexity of their models (including the number of parameters, FLOPs, and input shape).
If authors are participating in multiple tracks, the differences between models and/or training should be stated in the paper. The ranking in the final competition can be added to the camera-ready version if the paper is accepted. If no paper is submitted or if it is shorter than 3 pages, the team will be removed from the competition. Authors of accepted papers will also have a chance to present their work during the ICME 2025 conference’s grand challenge session.
Please use ICMEW2025 CMT website (opens in new tab) to submit your paper (https://cmt3.research.microsoft.com/ICMEW2025 (opens in new tab)).