Benchmark

[makise-kurisu-004]

Hi,

Thanks for this good work of MLLM-4D.

May I ask how we can obtain the benchmark (MLLM4DBench)? Is there a release schedule?

Thanks again.

[flow0314]

Hi kurisu, thanks for your interest in our work!
We have finished cleaning up the code and data, and have officially released our checkpoints, some datasets and codebase. Please note that MLLM4D-Bench consists of QA pairs similar to our MLLM4D-2M, but the videos are sourced from test sets. We plan to release the remaining materials of our project once the related paper completes the review process.
Thanks for your patience and understanding!