Semantic-Based Active Perception for Humanoid Visual Tasks with Foveal Sensors

arXiv:2404.10836v1 Announce Type: new
Abstract: The aim of this work is to establish how accurately a recent semantic-based foveal active perception model is able to complete visual tasks that are regularly performed by humans, namely, scene exploration and visual search. This model exploits the ability of current object detectors to localize and classify a large number of object classes and to update a semantic description of a scene across multiple fixations. It has been used previously in scene exploration tasks. …

abstract aim arxiv cs.cv current detectors eess.iv exploits exploration humanoid humans object perception search semantic sensors tasks type visual work