1
views
0
recommends
+1 Recommend
0 collections
    0
    shares
      • Record: found
      • Abstract: found
      • Article: found
      Is Open Access

      An End-to-End Network for Generating Social Relationship Graphs

      Preprint
      , ,

      Read this article at

      Bookmark
          There is no author summary for this article yet. Authors can add summaries to their articles on ScienceOpen to make them more accessible to a non-specialist audience.

          Abstract

          Socially-intelligent agents are of growing interest in artificial intelligence. To this end, we need systems that can understand social relationships in diverse social contexts. Inferring the social context in a given visual scene not only involves recognizing objects, but also demands a more in-depth understanding of the relationships and attributes of the people involved. To achieve this, one computational approach for representing human relationships and attributes is to use an explicit knowledge graph, which allows for high-level reasoning. We introduce a novel end-to-end-trainable neural network that is capable of generating a Social Relationship Graph - a structured, unified representation of social relationships and attributes - from a given input image. Our Social Relationship Graph Generation Network (SRG-GN) is the first to use memory cells like Gated Recurrent Units (GRUs) to iteratively update the social relationship states in a graph using scene and attribute context. The neural network exploits the recurrent connections among the GRUs to implement message passing between nodes and edges in the graph, and results in significant improvement over previous methods for social relationship recognition.

          Related collections

          Most cited references20

          • Record: found
          • Abstract: not found
          • Article: not found

          Dual-Process Models in Social and Cognitive Psychology: Conceptual Integration and Links to Underlying Memory Systems

            Bookmark
            • Record: found
            • Abstract: not found
            • Article: not found

            Visual Genome: Connecting Language and Vision Using Crowdsourced Dense Image Annotations

              Bookmark
              • Record: found
              • Abstract: found
              • Article: not found

              Role of facial expressions in social interactions.

              The expressions we see in the faces of others engage a number of different cognitive processes. Emotional expressions elicit rapid responses, which often imitate the emotion in the observed face. These effects can even occur for faces presented in such a way that the observer is not aware of them. We are also very good at explicitly recognizing and describing the emotion being expressed. A recent study, contrasting human and humanoid robot facial expressions, suggests that people can recognize the expressions made by the robot explicitly, but may not show the automatic, implicit response. The emotional expressions presented by faces are not simply reflexive, but also have a communicative component. For example, empathic expressions of pain are not simply a reflexive response to the sight of pain in another, since they are exaggerated when the empathizer knows he or she is being observed. It seems that we want people to know that we are empathic. Of especial importance among facial expressions are ostensive gestures such as the eyebrow flash, which indicate the intention to communicate. These gestures indicate, first, that the sender is to be trusted and, second, that any following signals are of importance to the receiver.
                Bookmark

                Author and article information

                Journal
                23 March 2019
                Article
                1903.09784
                7584f848-6aee-4cc4-aa4b-480caa4e9978

                http://creativecommons.org/licenses/by/4.0/

                Custom metadata
                CVPR 2019
                cs.CV

                Computer vision & Pattern recognition
                Computer vision & Pattern recognition

                Comments

                Comment on this article