What to Say and When to Say it: Live Fitness Coaching as a Testbed for Situated Interaction?

Qualcomm Exercise Video Dataset (QEVD): FIT-300k

Qualcomm Exercise Video Dataset (QEVD): FIT-COACH

QEVD-FIT-300k: IEEE LPCVC @ CVPR 2026

Link to Track 2 of the IEEE Low Power Computer Vision Challenge @ CVPR 2026.
An example video classification solution is available here.

FIT-COACH: VAR @ CVPR 2026

Link to AI Coach: Fitness challenge of the VAR Workshop @ CVPR 2026.
A quick start guide that implements a Qwen3-VL-2B-Instruct baseline is available here.
Official implementation of the Stream-VLM baseline is available here.

Citation

            
@inproceedings{fitcoach,
    title={What to Say and When to Say it: Live Fitness Coaching as a Testbed for Situated Interaction?},
    author={Panchal, S., Bhattacharyya, A., Berger, G., Mercier, A., Böhm, C., Dietrichkeit, F., Pourreza, R., Li, X., Maden, P., Lee, M., Todorovich, M., Bax, I., Memisevic, R.},
    booktitle={The Thirty-eight Annual Conference on Neural Information Processing Systems},
    year={2024},
}