HIGHLIGHTS
- What: This analysis demonstrates three advantages to this paper`s approach. The paper introduces a two-level image classification method to improve computer vision interpretability. The authors demonstrate this approach on a new dataset of 141,538 protest images from 10 countries. This paper focuses on the COCO and LVIS datasets.
- Who: Stefan Scholz et al. from the Center for Image Analysis in the Social Sciences, University of Konstanz, Konstanz, Germany have published the Article: Improving Computer Vision Interpretability: Transparent Two-Level Classification for Complex Scenes, in the Journal: (JOURNAL)
- How: This paper . . .
If you want to have access to all the content you need to log in!
Thanks :)
If you don't have an account, you can create one here.