BEDLAM
Emerging1papers using it
2026first seen
The 'BEDLAM' dataset is a benchmark used to evaluate human mesh recovery and tracking performance in multi-person scenes, particularly focusing on challenges such as occlusions and entity tracking over time.