Research map / Detection and segmentation / Open-set and few-shot detection
CountSE: Soft Exemplar Open-set Object Counting
Shuai Liu, Peng Zhang, Shiwei Zhang, Wei Ke
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 Semantic-guided Exemplar Selection (SES) (ablation) | GroundingDINO text-only baseline · FSC-147 val | MAE: 12.07 -> 8.56 | Improved |
| add SES (ablation) | GroundingDINO text-only baseline · FSC-147 test | MAE: 11.52 -> 8.53 | Improved |
| add SES (ablation) | GroundingDINO text-only baseline · FSC-147 val | RMSE: 57.02 -> 49.31 | Improved |
| add Clustering-based Exemplar Filtering (CEF) after SES (ablation) | CountSE · FSC-147 val | MAE: 8.56 -> 8.51 | Improved |
| add CEF after SES (ablation) | CountSE · FSC-147 test | MAE: 8.53 -> 7.84 | Improved |
| add CEF after SES (ablation) | CountSE · FSC-147 val | RMSE: 49.31 -> 54.93 | Hurt |
| set number of soft exemplars N to 18 instead of 14 (ablation) | CountSE · FSC-147 test | MAE: 8.53 -> 7.84 | Improved |
| increase N from 18 to 22 (ablation) | CountSE · FSC-147 test | MAE: 7.84 -> 9.24 | Hurt |
| set loss weight lambda to 5 instead of 3 (ablation) | CountSE · FSC-147 test | MAE: 9.59 -> 7.84 | Improved |
| increase lambda from 5 to 7 (ablation) | CountSE · FSC-147 test | MAE: 7.84 -> 9.38 | Hurt |
| use CountSE soft exemplars instead of GroundingREC text-only | CountSE · FSC-147 test | MAE: 10.12 -> 7.84 | Improved |
| use CountSE instead of CountGD (text) on CARPK | CountSE · CARPK | MAE: 3.83 -> 2.79 | Improved |
Closest papers by meaning
- CountGD: Multi-Modal Open-World Counting (NeurIPS 2024 · similarity 0.89)
- VLCounter: Text-Aware Visual Representation for Zero-Shot Object Counting (AAAI 2024 · similarity 0.88)
- T2ICount: Enhancing Cross-modal Understanding for Zero-Shot Counting (CVPR 2025 · similarity 0.88)
- DAVE - A Detect-and-Verify Paradigm for Low-Shot Counting (CVPR 2024 · similarity 0.87)
- OmniCount: Multi-label Object Counting with Semantic-Geometric Priors (AAAI 2025 · similarity 0.87)
- A Novel Unified Architecture for Low-Shot Counting by Detection and Segmentation (NeurIPS 2024 · similarity 0.86)
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.
