Abstract: With the rapid growth of automated retail and smart supermarkets, commodity package recognition faces challenges like complex backgrounds, multi-scale targets, and dense occlusion. To ...
aDepartment of Surgery, Regional Academic Cancer Center Utrecht, UMC Utrecht Cancer Center & St Antonius Hospital Nieuwegein, Utrecht University, Utrecht, Netherlands bDepartment of Medical Oncology, ...
Abstract: The recognition of easily confused activities is a key issue that limits the performance improvement of traditional human activity recognition (HAR) algorithms. To address this, this paper ...