HA-ViD paper accepted at NeurIPS 2023

We are pleased to announce the acceptance of our paper, “HA-ViD: A Human Assembly Video Dataset for Comprehensive Assembly Knowledge Understanding” at the esteemed NeurIPS conference. This achievement marks a milestone in our journey toward understanding the comprehensive knowledge of assembly applications such as human-robot collaboration, robot skill learning and assembly process monitoring.

HA-ViD captures the essence of industrial human assembly scenarios through the designed Generic Assembly Box, 3-stage progressive assembly setup and consistent human-robot shared annotations. The applications of this work enables research towards understanding natural human behaviours, progressive knowledge acquisition and two-hand collaboration patterns during assembly.

For more information on HA-ViD, please visit our project website