Research map / Architectures and efficiency / CNN design
Explicit Spatial Encoding for Deep Local Descriptors
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 combined Cartesian and polar explicit spatial encoding with separate convolutional parts at s=2 instead of HardNet+'s fully connected layer | Deep local descriptor · PhotoTourism | FPR@95 (mean): 1.43 -> 1.36 | Helped |
| Use combined Cartesian and polar explicit spatial encoding with separate convolutional parts at s=1 instead of HardNet+'s fully connected layer | Deep local descriptor · PhotoTourism | FPR@95 (mean): 1.43 -> 1.53 | Hurt |
| Remove all spatial encoding and use spatial sum pooling (ablation) | Proposed deep local descriptor · HPatches | mAP | Hurt |
| Replace tolerant explicit position encoding with rigid spatial concatenation of the feature map (ablation) | Proposed deep local descriptor · HPatches | mAP | Hurt |
| Add a learned matrix M_sum after spatial sum pooling (ablation) | Spatial-sum descriptor · HPatches | mAP | No clear effect |
Closest papers by meaning
- L2-Net: Deep Learning of Discriminative Patch Descriptor in Euclidean Space (CVPR 2017 · similarity 0.89)
- Kernelized Subspace Pooling for Deep Local Descriptors (CVPR 2018 · similarity 0.86)
- Local Descriptors Optimized for Average Precision (CVPR 2018 · similarity 0.86)
- Local Features and Visual Words Emerge in Activations (CVPR 2019 · similarity 0.86)
- An Image is Worth More Than 16x16 Patches: Exploring Transformers on Individual Pixels (ICLR 2025 · similarity 0.83)
- Learning Input Encodings for Kernel-Optimal Implicit Neural Representations (ICML 2025 · 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.