Research map / Detection and segmentation / CNN features
Learnable Tree Filter for Structure-preserving Feature Transform
Lin Song, Yanwei Li, Zeming Li, Gang Yu, Hongbin Sun, Jian Sun, Nanning Zheng
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 tree filtering modules at stages 1-3 (ablation) | ResNet-50 decoder · PASCAL VOC 2012 val | mIoU (%): 70.2 -> 74.7 | Improved |
| add tree filtering modules at stages 1-3 (ablation) | ResNet-101 decoder · PASCAL VOC 2012 val | mIoU (%): 75.0 -> 77.1 | Improved |
| increase tree-weight groups from 1 to 16 (ablation) | single-stage tree filter · PASCAL VOC 2012 val | mIoU (%): 72.1 -> 74.4 | Improved |
| increase groups from 16 to 32 (ablation) | single-stage tree filter · PASCAL VOC 2012 val | mIoU (%): 74.4 -> 74.4 | No clear effect |
| replace PSP block with tree filter (OS 8) (ablation) | ResNet-50 · PASCAL VOC 2012 val | mIoU (%): 74.3 -> 74.9 | Improved |
| add tree filter to PSP block (ablation) | ResNet-50 · PASCAL VOC 2012 val | mIoU (%): 74.3 -> 75.4 | Improved |
| add non-local block on top of tree filter (ablation) | tree filter · PASCAL VOC 2012 val | mIoU (%): 74.9 -> 74.9 | No clear effect |
| add tree filter with decoder (OS 32) (ablation) | ResNet-50 · PASCAL VOC 2012 val | mIoU (%): 67.3 -> 72.5 | Improved |
| add multi-stage tree filter to decoder with extra components (ablation) | ResNet-101 · PASCAL VOC 2012 val | mIoU (%): 78.3 -> 79.4 | Improved |
| add extra GAP and ResBlock components (ablation) | ResNet-50 decoder · PASCAL VOC 2012 val | mIoU (%): 67.3 -> 72.5 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- PSANet — PSANet: Point-wise Spatial Attention Network for Scene Parsing (ECCV 2018) · better on 1 of 1 shared result
- ExFuse — ExFuse: Enhancing Feature Fusion for Semantic Segmentation (ECCV 2018) · better on 1 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Rethinking Learnable Tree Filter for Generic Feature Transform (NeurIPS 2020 · similarity 0.95)
- Semantic Image Segmentation With Task-Specific Edge Detection Using CNNs and a Discriminatively Trained Domain Transform (CVPR 2016 · similarity 0.86)
- Deep Edge Filter: Return of the Human-Crafted Layer in Deep Learning (NeurIPS 2025 · similarity 0.85)
- Tree Energy Loss: Towards Sparsely Annotated Semantic Segmentation (CVPR 2022 · similarity 0.85)
- ViTree: Single-Path Neural Tree for Step-Wise Interpretable Fine-Grained Visual Categorization (AAAI 2024 · similarity 0.84)
- Learning Hierarchical Image Segmentation For Recognition and By Recognition (ICLR 2024 · similarity 0.84)
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.
