Research map / Detection and segmentation / Weakly supervised detection and segmentation
Leveraging Auxiliary Tasks With Affinity Learning for Weakly Supervised Semantic Segmentation
Lian Xu, Wanli Ouyang, Mohammed Bennamoun, Farid Boussaid, Ferdous Sohel, Dan Xu
Read the paperCheck your idea against 137,217 papers
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 |
|---|---|---|---|
| add multi-label image classification auxiliary task (ablation) | segmentation baseline · PASCAL VOC 2012 val | mIoU (%): 56.9 -> 57.6 | Improved |
| add saliency detection auxiliary task (ablation) | segmentation baseline · PASCAL VOC 2012 val | mIoU (%): 56.9 -> 59.8 | Improved |
| add both classification and saliency auxiliary tasks (ablation) | segmentation baseline · PASCAL VOC 2012 val | mIoU (%): 56.9 -> 60.8 | Improved |
| add segmentation-only affinity learning (ablation) | AuxSegNet multi-task baseline · PASCAL VOC 2012 val | mIoU (%): 60.8 -> 61.5 | Improved |
| use cross-task affinity instead of segmentation-only affinity (ablation) | AuxSegNet · PASCAL VOC 2012 val | mIoU (%): 61.5 -> 62.6 | Improved |
| add saliency constraint on cross-task affinity (ablation) | AuxSegNet · PASCAL VOC 2012 val | mIoU (%): 62.6 -> 64.1 | Improved |
| iterative pseudo-label updating over stages (ablation) | AuxSegNet · PASCAL VOC 2012 val | mIoU (%): 64.1 -> 66.1 | Improved |
| refine CAM with cross-task affinity in first label update (ablation) | AuxSegNet pseudo labels · PASCAL VOC 2012 train (pseudo labels) | precision/recall/mIoU (%): recall 42.9 -> +8.9, mIoU 40.6 -> +8.6 | Improved |
| add CRF post-processing and multi-scale testing (ablation) | AuxSegNet · PASCAL VOC 2012 val | mIoU (%): 66.1 -> 69.0 | Improved |
| swap pre-trained saliency model to Zhang et al. (weakly supervised) (ablation) | AuxSegNet · PASCAL VOC 2012 val (pre-trained saliency model comparison) | mIoU (%): 53.0 -> 66.5 | Improved |
| swap pre-trained saliency model to PoolNet (ablation) | AuxSegNet · PASCAL VOC 2012 val (pre-trained saliency model comparison) | mIoU (%): 55.7 -> 68.4 | Improved |
| swap pre-trained saliency model to MINet (ablation) | AuxSegNet · PASCAL VOC 2012 val (pre-trained saliency model comparison) | mIoU (%): 55.8 -> 68.9 | Improved |
Closest papers by meaning
- Learning Affinity From Attention: End-to-End Weakly-Supervised Semantic Segmentation With Transformers (CVPR 2022 · similarity 0.92)
- CIAN: Cross-Image Affinity Net for Weakly Supervised Semantic Segmentation (AAAI 2020 · similarity 0.91)
- Label-efficient Segmentation via Affinity Propagation (NeurIPS 2023 · similarity 0.90)
- Learning Pixel-Level Semantic Affinity With Image-Level Supervision for Weakly Supervised Semantic Segmentation (CVPR 2018 · similarity 0.90)
- Weakly-Supervised Semantic Segmentation by Iterative Affinity Learning (IJCV 2020 · similarity 0.90)
- W2P: Switching from Weak Supervision to Partial Supervision for Semantic Segmentation (AAAI 2024 · similarity 0.88)
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.
