Research map / Detection and segmentation / Weakly supervised detection and segmentation
Adaptive Affinity Fields for Semantic Segmentation
Tsung-Wei Ke, Jyh-Jing Hwang, Ziwei Liu, Stella X. Yu
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 AAF training loss | PSPNet · Cityscapes val | mIoU: 76.72 -> 79.24 | Improved |
| add AAF training loss | PSPNet · PASCAL VOC 2012 val | mIoU: 80.12 -> 82.39 | Improved |
| add AAF training loss | FCN · PASCAL VOC 2012 val | mIoU: 68.91 -> 71.95 | Improved |
| add AAF training loss | FCN · Cityscapes val | mIoU: 66.77 -> 69.14 | Improved |
| add AAF training loss | PSPNet · PASCAL VOC 2012 test | mIoU: 80.63 -> 82.17 | Improved |
| add AAF training loss | PSPNet · Cityscapes test | mIoU: 76.30 -> 79.07 | Improved |
| add adversarially learned adaptive field size to fixed affinity fields (ablation) | PSPNet · PASCAL VOC 2012 val | mIoU: 81.80 -> 82.39 | Improved |
| add adversarially learned adaptive field size to fixed affinity fields (ablation) | PSPNet · Cityscapes val | mIoU: 78.72 -> 79.24 | Improved |
| add adversarial learning of field size to fixed affinity fields (ablation) | FCN · PASCAL VOC 2012 val | mIoU: 71.07 -> 71.95 | Improved |
| replace GAN structural prior with AAF | PSPNet · Cityscapes val | mIoU: 76.20 -> 79.24 | Improved |
| replace CRF structural prior with AAF | PSPNet · Cityscapes val | mIoU: 76.53 -> 79.24 | Improved |
| add 5x5 kernel to 3x3 kernel affinity fields (ablation) | AAF · PASCAL VOC 2012 val (kernel size study) | mIoU: 79.00 -> 79.45 | Improved |
Closest papers by meaning
- Learning Pixel-Level Semantic Affinity With Image-Level Supervision for Weakly Supervised Semantic Segmentation (CVPR 2018 · similarity 0.91)
- Learning to Adapt Structured Output Space for Semantic Segmentation (CVPR 2018 · similarity 0.88)
- Semi Supervised Semantic Segmentation Using Generative Adversarial Network (ICCV 2017 · similarity 0.88)
- Learning Affinity From Attention: End-to-End Weakly-Supervised Semantic Segmentation With Transformers (CVPR 2022 · similarity 0.88)
- Few-Shot Semantic Image Synthesis With Class Affinity Transfer (CVPR 2023 · similarity 0.88)
- Repurposing GANs for One-Shot Semantic Part Segmentation (CVPR 2021 · 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.
