4
views
0
recommends
+1 Recommend
0 collections
    0
    shares
      • Record: found
      • Abstract: found
      • Article: found
      Is Open Access

      Active query-driven visual search using probabilistic bisection and convolutional neural networks

      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

          We present a novel efficient object detection and localization framework based on the probabilistic bisection algorithm. A convolutional neural network is trained and used as a noisy oracle that provides answers to input query images. The responses along with error probability estimates obtained from the CNN are used to update beliefs on the object location along each dimension. We show that querying along each dimension achieves the same lower bound on localization error as the joint query design. Finally, we provide experimental results on a face localization task that showcase the effectiveness of our approach in comparison to sliding window techniques.

          Related collections

          Most cited references8

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

          Recent advances in convolutional neural networks

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

            Sequential transmission using noiseless feedback

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

              Bisection Search with Noisy Responses

                Bookmark

                Author and article information

                Journal
                28 June 2018
                Article
                1806.11223
                17758e0f-1503-48f7-a106-8bf2f7395d80

                http://arxiv.org/licenses/nonexclusive-distrib/1.0/

                History
                Custom metadata
                5 pages, 4 figures, 1 table, submitted to GlobalSIP 2018
                cs.CV

                Computer vision & Pattern recognition
                Computer vision & Pattern recognition

                Comments

                Comment on this article