Research map / Label-efficient and robust learning / Recognition and long-tail
Encoding Spatial Distribution of Convolutional Features for Texture Representation
Yong Xu, Feng Li, Zhile Chen, Jinxiu Liang, Yuhui Quan
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 FE module (with BLP) to FENet (ablation) | FENet ResNet-18 · GTOS-M | accuracy (%): 82.29 -> 85.10 | Improved |
| add FE module (with BLP) to FENet (ablation) | FENet ResNet-18 · KTH | accuracy (%): 83.25 -> 86.62 | Improved |
| add FE module (with BLP) to FENet (ablation) | FENet ResNet-50 · KTH | accuracy (%): 83.82 -> 88.24 | Improved |
| add FE module (with BLP) to FENet (ablation) | FENet ResNet-50 · GTOS-M | accuracy (%): 82.57 -> 85.20 | Improved |
| add GAP layer in FE (ablation) | FENet ResNet-18 · GTOS-M | accuracy (%): 83.86 -> 85.10 | Improved |
| add GAP layer in FE (ablation) | FENet ResNet-18 · KTH | accuracy (%): 86.43 -> 86.62 | Improved |
| add FE module to DeepTEN (ablation) | DeepTEN ResNet-50 · KTH | accuracy (%): 82.00 ± 3.30 -> 86.93 ± 0.30 | Improved |
| add FE module to DeepTEN (ablation) | DeepTEN ResNet-50 · FMD | accuracy (%): 80.20 ± 0.90 -> 86.32 ± 0.43 | Improved |
| add FE module to DeepTEN (ablation) | DeepTEN ResNet-50 · MINC | accuracy (%): 81.3 -> 83.62 ± 0.11 | Improved |
| use FENet instead of DSRNet/MAPNet (ResNet-18) | FENet · DTD | accuracy (%): 69.59 ± 0.04 vs DSRNet 71.2 ± 0.7 | Mixed |
| use FENet instead of DSRNet (ResNet-50) | FENet · GTOS-M | accuracy (%): 85.20 ± 0.43 vs DSRNet 87.03 ± 1.5 | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- DeepTEN — Deep TEN: Texture Encoding Network (CVPR 2017) · better on 4 of 4 shared results
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
- Fractal-Domain Vision Graph Neural Network for Remote Sensing Ground Target Classification (TPAMI 2026 · similarity 0.88)
- Fractal Dimension Invariant Filtering and Its CNN-Based Implementation (CVPR 2017 · similarity 0.87)
- Learning Fractals by Gradient Descent (AAAI 2023 · similarity 0.86)
- Visualizing and Understanding Deep Texture Representations (CVPR 2016 · similarity 0.86)
- Deep TEN: Texture Encoding Network (CVPR 2017 · similarity 0.85)
- Deep Texture Recognition via Exploiting Cross-Layer Statistical Self-Similarity (CVPR 2021 · 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.
