Research map / Architectures and efficiency / RNNs and normalization
Hyperdimensional Computing as a Framework for Systematic Aggregation of Image 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 |
|---|---|---|---|
| aggregate DELF local descriptors with binding to encoded image positions | DELF-HDC · 23 place-recognition sequence comparisons | average precision: +20% to runner-up on average; 3.6x runner-up in the worst case | Helped |
| add position binding before bundling descriptors (ablation) | DELF-HDC · Nordland spring-summer matching experiment | similarity to a true matching feature | Helped |
| use HDC bundling instead of concatenating projected holistic descriptors (ablation) | multiple holistic descriptors · all datasets in the place-recognition evaluation | average precision | No clear effect |
| reduce the HDC representation dimension (ablation) | DELF-HDC · all evaluated datasets | average precision: >=512 dimensions | Mixed |
| increase the number of local features (ablation) | DELF-HDC · GardensPointWalking | average precision | Helped |
| set horizontal position parameter n_x to 2 (ablation) | DELF-HDC · GardensPointWalking Day Left - Night Right | average precision: AP=0.59 (+0.13) | Helped |
| increase spatial parameters n_x and n_y (ablation) | DELF-HDC · GardensPointWalking, small viewpoint changes | average precision | Helped |
| use lower n_x to accommodate larger horizontal viewpoint changes (ablation) | DELF-HDC · GardensPointWalking left-right sequence comparisons | average precision | Helped |
| mean-center descriptors during preprocessing | image descriptors · place recognition | separation of true and false match distributions | Helped |
Closest papers by meaning
- Towards Implicit Aggregation: Robust Image Representation for Place Recognition in the Transformer Era (NeurIPS 2025 · similarity 0.85)
- Bridging the Gap Between Hyperdimensional Computing and Kernel Methods via the Nyström Method (AAAI 2025 · similarity 0.85)
- SuperVLAD: Compact and Robust Image Descriptors for Visual Place Recognition (NeurIPS 2024 · similarity 0.84)
- Optimal Transport Aggregation for Visual Place Recognition (CVPR 2024 · similarity 0.84)
- Neural Clustering based Visual Representation Learning (CVPR 2024 · similarity 0.82)
- Implicit Zoo: A Large-Scale Dataset of Neural Implicit Functions for 2D Images and 3D Scenes (NeurIPS 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.