Research map / Label-efficient and robust learning / Recognition and long-tail
Deep Texture Manifold for Ground Terrain Recognition
Jia Xue, Hang Zhang, Kristin Dana
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 |
|---|---|---|---|
| Combine orderless texture encoding with global-average-pooled spatial information using a bilinear model | ResNet-18 · GTOS-mobile, single scale | accuracy: 74.22 -> 76.07 | Improved |
| Combine orderless texture encoding with global-average-pooled spatial information using a bilinear model | ResNet-18 · GTOS-mobile, multi scale | accuracy: 76.12 -> 82.18 | Improved |
| Use multi-scale rather than single-scale training | DEP · GTOS-mobile | accuracy: 76.07 -> 82.18 | Improved |
| Use multi-scale rather than single-scale training | ResNet · GTOS-mobile | accuracy: 70.82 -> 73.16 | Improved |
| Use multi-scale rather than single-scale training | Bilinear CNN · GTOS-mobile | accuracy: 72.03 -> 75.43 | Improved |
| Use multi-scale rather than single-scale training | Deep-TEN · GTOS-mobile | accuracy: 74.22 -> 76.12 | Improved |
| Combine texture and spatial features in DEP instead of using global-average-pooled spatial features alone | ResNet-18 · GTOS-mobile, multi scale | accuracy: 73.16 -> 82.18; 9.02% | Improved |
| Combine texture and spatial features in DEP instead of texture details alone | Deep-TEN · GTOS-mobile, multi scale | accuracy: 76.12 -> 82.18; 6% | Improved |
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.
- Deep-TEN — 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
- Differential Viewpoints for Ground Terrain Material Recognition (TPAMI 2020 · similarity 0.87)
- TextureNet: Consistent Local Parametrizations for Learning From High-Resolution Signals on Meshes (CVPR 2019 · similarity 0.85)
- Deep TEN: Texture Encoding Network (CVPR 2017 · similarity 0.84)
- Patched Line Segment Learning for Vector Road Mapping (AAAI 2024 · similarity 0.84)
- From-Ground-To-Objects: Coarse-to-Fine Self-supervised Monocular Depth Estimation of Dynamic Objects with Ground Contact Prior (CVPR 2024 · similarity 0.83)
- DCV2I: A Practical Approach for Supporting Geographers’ Visual Interpretation in Dune Segmentation with Deep Vision Models (AAAI 2024 · similarity 0.82)
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.
