2020.12.18
Best Paper Honorable Mention Award at ACCV2020
- Technologies
- Machine learning
- National Institute of Advanced Industrial Science and Technology, Artificial Intelligence Research Center (AIRC)
- https://hirokatsukataoka16.github.io/Pretraining-without-Natural-Images/
At the Asian Conference on Computer Vision (ACCV) 2020, an international conference on computer vision, a paper presented by Kataoka et al. at the Artificial Intelligence Research Center, AIST, received the Best Paper Honorable Mention Award.
The paper proposes a novel concept, Formula-driven Supervised Learning. The paper shows the proposed Fractal DataBase (FractalDB), a database without natural images, is useful for pre-training.
Title: Pre-training without Natural Images
Authors: Hirokatsu Kataoka, Kazushige Okayasu, Asato Matsumoto, Eisuke Yamagata, Ryosuke Yamada, Nakamasa Inoue, Akio Nakamura, Yutaka Satoh
Technologies
1