@InProceedings{BlangerHiraJian:2021:ReNeBo,
               author = "Blanger, Leonardo and Hirata, Nina S. T. and Jiang, Xiaoyi",
          affiliation = "{University of Sao Paulo } and {University of Sao Paulo } and 
                         {University of M{\"u}nster}",
                title = "Reducing the need for bounding box annotations in Object Detection 
                         using Image Classification data",
            booktitle = "Proceedings...",
                 year = "2021",
               editor = "Paiva, Afonso and Menotti, David and Baranoski, Gladimir V. G. and 
                         Proen{\c{c}}a, Hugo Pedro and Junior, Antonio Lopes Apolinario 
                         and Papa, Jo{\~a}o Paulo and Pagliosa, Paulo and dos Santos, 
                         Thiago Oliveira and e S{\'a}, Asla Medeiros and da Silveira, 
                         Thiago Lopes Trugillo and Brazil, Emilio Vital and Ponti, Moacir 
                         A. and Fernandes, Leandro A. F. and Avila, Sandra",
         organization = "Conference on Graphics, Patterns and Images, 34. (SIBGRAPI)",
            publisher = "IEEE Computer Society",
              address = "Los Alamitos",
             keywords = "sample synthesis, object detection, pretraining, deep learning.",
             abstract = "We address the problem of training Object Detection models using 
                         significantly less bounding box annotated images. For that, we 
                         take advantage of cheaper and more abundant image classification 
                         data. Our proposal consists in automatically generating artificial 
                         detection samples, with no need of expensive detection level 
                         supervision, using images with classification labels only. We also 
                         detail a pretraining initialization strategy for detection 
                         architectures using these artificially synthesized samples, before 
                         finetuning on real detection data, and experimentally show how 
                         this consistently leads to more data efficient models. With the 
                         proposed approach, we were able to effectively use only 
                         classification data to improve results on the harder and more 
                         supervision hungry object detection problem. We achieve results 
                         equivalent to those of the full data scenario using only a small 
                         fraction of the original detection data for Face, Bird, and Car 
                         detection.",
  conference-location = "Gramado, RS, Brazil (virtual)",
      conference-year = "18-22 Oct. 2021",
                  doi = "10.1109/SIBGRAPI54419.2021.00035",
                  url = "http://dx.doi.org/10.1109/SIBGRAPI54419.2021.00035",
             language = "en",
                  ibi = "8JMKD3MGPEW34M/45C6UK8",
                  url = "http://urlib.net/ibi/8JMKD3MGPEW34M/45C6UK8",
           targetfile = "29.pdf",
        urlaccessdate = "2024, May 06"
}