Weakly Supervised Person Dataset (WSPD)

  • Our weak-supervision surpassed a supervised pre-training.

    To construct an algorithm that can provide robust person detection, we present a dataset with over 8 million images that was produced in a weakly supervised manner. Through labor-intensive human annotation, the person detection research community has produced relatively small datasets containing on the order of 100,000 images, such as the EuroCity Persons dataset, which includes 240,000 bounding boxes. Therefore, we have collected 8.7 million images of persons based on a two-step collection process, namely person detection with an existing detector and data refinement for false positive suppression. According to the experimental results, the Weakly Supervised Person Dataset (WSPD) is simple yet effective for person detection pre-training. In the context of pre-trained person detection algorithms, our WSPD pre-trained model has 13.38 and 6.38% better accuracy than the same model trained on the fully supervised ImageNet and EuroCity Persons datasets, respectively, when verified with the Caltech Pedestrian.

  • Post destination

    arXiv pre-print, ViEW 2019 Best Student Paper

  • Member

    Munetaka Minoguchi, Ken Okayama, Yutaka Satoh, Hirokatsu Kataoka (AIST)