SynthGait-19K: A Physically Grounded Synthetic Video Dataset for Gait Estimation
uoft · hf · 2026-09-09
The uoft team released SynthGait-19k on Hugging Face, a large-scale synthetic video dataset for gait analysis built with physically grounded simulation.
The dataset enables benchmarking of video-based gait parameter estimation and demonstrates that synthetic supervision transfers to real data, offering a practical path for training and evaluating gait estimation where real-world annotation is hard.
More from Research
- Open-source 4B VLM RL-trained to play GeoGuessr beats GPT 5.4 mini and Haiku on one A100 — SergioPaniego · 2026-09-09
- Granularity and the 'birdiest bird': self-supervised clusters cut across ImageNet labels — y_m_asano · 2026-09-09
- Reviewer finds AI-written peer reviews rampant at ARR, calls for benchmarks of AI reviews — ChenhaoTan · 2026-09-09
- MirroS finds S-Space: a manipulable spatial workspace inside multimodal models — HeyAmit_ · 2026-09-09
- Synthetic Data May Leak Answers, Security Expert Questions — matthew_d_green · 2026-09-09
- AI Math Gains Don't Solve Open-Ended Discovery, Says Researcher — kenneth0stanley · 2026-09-09