Research map / Detection and segmentation / Weakly supervised detection and segmentation
Full or Weak Annotations? An Adaptive Strategy for Budget-Constrained Annotation Campaigns
Javier Gamazo Tejero, Martin S. Zinkernagel, Sebastian Wolf, Raphael Sznitman, Pablo Márquez-Neila
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Use adaptive rather than fixed annotation-budget allocation | Segmentation annotation campaigns · Cityscapes | relative performance (%): 73.9 -> 73.4 | Hurt |
| Use adaptive rather than fixed annotation-budget allocation | Segmentation annotation campaigns · OCT | relative performance (%): 95.4 -> 92.6 | Hurt |
| Use adaptive rather than fixed annotation-budget allocation | Segmentation annotation campaigns · SUIM | relative performance (%): 81.3 -> 81.9 | Improved |
| Use adaptive rather than fixed annotation-budget allocation | Segmentation annotation campaigns · VOC | relative performance (%): 46.0 -> 46.0 | No clear effect |
| Use adaptive rather than fixed annotation-budget allocation | Segmentation annotation campaigns · Average | relative performance (%): 73.0 -> 73.5 | Improved |
| Vary the segmentation annotation cost assumption alpha_s (ablation) | Adaptive allocation method · Cityscapes | relative performance (%) | No clear effect |
| Increase the number of adaptive steps T from low values (ablation) | Adaptive allocation method · OCT and SUIM | strategy reliability | Improved |
| Increase the number of adaptive steps T (ablation) | Adaptive allocation method · OCT, SUIM, Cityscapes, and VOC | strategy variance | Improved |
Closest papers by meaning
- Adaptive Early-Learning Correction for Segmentation From Noisy Annotations (CVPR 2022 · similarity 0.89)
- Block Annotation: Better Image Annotation With Sub-Image Decomposition (ICCV 2019 · similarity 0.88)
- Cut out the annotator, keep the cutout: better segmentation with weak supervision (ICLR 2021 · similarity 0.88)
- Scaling Up Instance Annotation via Label Propagation (ICCV 2021 · similarity 0.87)
- Point Cloud Semantic Segmentation with Sparse and Inhomogeneous Annotations (AAAI 2025 · similarity 0.86)
- Low-Biased General Annotated Dataset Generation (CVPR 2025 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
