LOOC: Localize Overlapping Objects with Count Supervision
This addresses the challenge of reducing annotation costs for object localization in dense scenes, though it is incremental as it builds on existing counting methods.
The paper tackles the problem of localizing objects in dense scenes using only count supervision, which requires less human effort than point-level or bounding box annotations, and proposes LOOC, a method that achieves a strong new baseline for localization and outperforms state-of-the-art methods for counting under this supervision.
Acquiring count annotations generally requires less human effort than point-level and bounding box annotations. Thus, we propose the novel problem setup of localizing objects in dense scenes under this weaker supervision. We propose LOOC, a method to Localize Overlapping Objects with Count supervision. We train LOOC by alternating between two stages. In the first stage, LOOC learns to generate pseudo point-level annotations in a semi-supervised manner. In the second stage, LOOC uses a fully-supervised localization method that trains on these pseudo labels. The localization method is used to progressively improve the quality of the pseudo labels. We conducted experiments on popular counting datasets. For localization, LOOC achieves a strong new baseline in the novel problem setup where only count supervision is available. For counting, LOOC outperforms current state-of-the-art methods that only use count as their supervision. Code is available at: https://github.com/ElementAI/looc.