Research map / Detection and segmentation / Open-set and few-shot detection
DualCount: Structurally Consistent Density and Point Modeling for Zero-Shot Object Counting
Xuan Cuong Ngo
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 center-of-mass alignment loss (ablation) | DualCount · FSC-147 (zero-shot) test | MAE: 12.84 -> 11.32 | Improved |
| add per-instance mass conservation loss (ablation) | DualCount · FSC-147 (zero-shot) test | MAE: 13.05 -> 11.32 | Improved |
| add count consistency loss (ablation) | DualCount · FSC-147 (zero-shot) test | MAE: 14.25 -> 11.32 | Improved |
| add contrastive loss (ablation) | DualCount · FSC-147 (zero-shot) test | MAE: 14.49 -> 11.32 | Improved |
| add point decoder to density-only model (ablation) | DualCount · FSC-147 (zero-shot) test | MAE: 14.12 -> 11.32 | Improved |
| add density decoder to point-only model (ablation) | DualCount · FSC-147 (zero-shot) test | MAE: 14.54 -> 11.32 | Improved |
| add feature enhancer (ablation) | DualCount · FSC-147 (zero-shot) test | MAE: 13.11 -> 11.32 | Improved |
| replace CLIP backbone with SwinT+BERT (ablation) | DualCount · FSC-147 (zero-shot) test | MAE: 12.33 -> 11.32 | Improved |
| add all loss components (RMSE) (ablation) | DualCount · FSC-147 (zero-shot) test | RMSE: 100.45 -> 97.16 | Improved |
| apply DualCount vs CountGD on PUCPR+ | DualCount · PUCPR+ (zero-shot) | MAE: 24.31 -> 24.27 | Improved |
| apply DualCount vs CountGD on CARPK | DualCount · CARPK (zero-shot) | MAE: 3.83 -> 7.95 | 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.
- DAVE — DAVE - A Detect-and-Verify Paradigm for Low-Shot Counting (CVPR 2024) · 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
- VLCounter: Text-Aware Visual Representation for Zero-Shot Object Counting (AAAI 2024 · similarity 0.90)
- DAVE - A Detect-and-Verify Paradigm for Low-Shot Counting (CVPR 2024 · similarity 0.89)
- Point Segment and Count: A Generalized Framework for Object Counting (CVPR 2024 · similarity 0.89)
- T2ICount: Enhancing Cross-modal Understanding for Zero-Shot Counting (CVPR 2025 · similarity 0.89)
- A Novel Unified Architecture for Low-Shot Counting by Detection and Segmentation (NeurIPS 2024 · similarity 0.88)
- OmniCount: Multi-label Object Counting with Semantic-Geometric Priors (AAAI 2025 · similarity 0.87)
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.
