Researchers have introduced DGHMesh, a large-scale dataset and benchmark designed to improve human mesh reconstruction using millimeter-wave (mmWave) radar. The dataset includes synchronized data from radar, RGB images, and 3D annotations for 15 subjects performing various actions. It is specifically designed to test the generalization capabilities of reconstruction methods across different configurations, such as shifts in human position and orientation. AI
Summary written by gemini-2.5-flash-lite from 1 source. How we write summaries →
IMPACT Provides a new benchmark for evaluating and improving human mesh reconstruction algorithms using radar data.
RANK_REASON This is a research paper introducing a new dataset and benchmark for a specific AI task.