Research map / Architectures and efficiency / RNNs and normalization
Sparse Coding for Third-Order Super-Symmetric Tensor Descriptors With Application to Texture Recognition
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Use third-order rather than second-order global image pooling | Global image descriptor · PASCAL VOC07 | accuracy: 54.0% -> 62.7% | Helped |
| Use third-order rather than second-order global image pooling | Global image descriptor · Caltech 101 | accuracy: 78.5% -> 83.9% | Helped |
| Use third-order rather than second-order global image pooling | Global image descriptor · Flowers 102 | accuracy: 83.1% -> 89.0% | Helped |
| Replace linear feature representation with RBF features (ablation) | TOSST texture descriptor · Brodatz | accuracy: 93.9±0.2% -> 99.4±0.2% | Helped |
| Add luminance to the linear feature descriptor (ablation) | TOSST texture descriptor · Brodatz | accuracy: 93.9±0.2% -> 99.4±0.1% | Helped |
| Add luminance to the RBF feature descriptor (ablation) | TOSST texture descriptor · Brodatz | accuracy: 99.4±0.2% -> 99.9±0.08% | No clear effect |
| Compress the third-order global image descriptor with a learned dictionary | PASCAL VOC07 image signatures · PASCAL VOC07 | MAP: 61.3% MAP -> 61.2% MAP | No clear effect |
| Use a learned dictionary rather than a random dictionary (ablation) | Sparse signature compression · PASCAL VOC07 | MAP: about 4.3% worse | Helped |
| Combine dictionary learning with Product Quantization (ablation) | Compressed image signatures · PASCAL VOC07 | MAP: up to 3% MAP | Helped |
| Impose low-rank constraints on dictionary atoms (ablation) | Tensor dictionary learning · Brodatz | accuracy: up to 2% | Mixed |
| Use low sparsity in the coding and pooling steps (ablation) | Tensor sparse coding · Brodatz | accuracy | Helped |
Closest papers by meaning
- Robust High-Order Tensor Compressive Sensing Based on M-Estimators (AAAI 2026 · similarity 0.85)
- Distribution-Aware Tensor Decomposition for Compression of Convolutional Neural Networks (NeurIPS 2025 · similarity 0.85)
- Nonparametric Decomposition of Sparse Tensors (ICML 2021 · similarity 0.85)
- Efficient Multi-Dimensional Tensor Sparse Coding Using t-Linear Combination (AAAI 2018 · similarity 0.85)
- Coarse-To-Fine Tensor Trains for Compact Visual Representations (ICML 2024 · similarity 0.85)
- Performance Gaps in Multi-view Clustering under the Nested Matrix-Tensor Model (ICLR 2024 · similarity 0.83)
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.