Research map / Vision-language and multimodal / CLIP and zero-shot
HER-Count: Learning Hyper-Exemplar Representation for Generalized Zero-Shot Object Counting
Jianing Li, Xiaobin Liu, Ruihan Xu
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 hierarchical fusion (HFS) of HERs into multiple encoder layers (ablation) | HER · FSC147 val | MAE: 16.85 -> 15.53 | Improved |
| add global discrimination enhancement (GDE) (ablation) | HER · FSC147 val | MAE: 16.85 -> 14.33 | Improved |
| use both image- and class-level GDE instead of image-level only (ablation) | HER + GDE · FSC147 val | MAE: 15.18 -> 14.33 | Improved |
| combine HFS and GDE (ablation) | HER · FSC147 test | MAE: 17.17 -> 14.23 | Improved |
| synthesize HERs from image and class name rather than image only (ablation) | HER-Count · FSC147 val | MAE: 16.55 -> 13.56 | Improved |
| synthesize HERs from image and class name rather than text only (ablation) | HER-Count · FSC147 val | MAE: 15.77 -> 13.56 | Improved |
| vary LMD threshold from 0.01 to 0.2 (ablation) | HER-Count · FSC147 val | MAE: 13.56 at 0.05; 13.89 at 0.01; 13.92 at 0.2 | No clear effect |
| use Qwen3-VL-2B MLLM instead of CLIP ViT-L/14 (ablation) | HER-Count · FSC147 val | MAE: 14.73 -> 13.56 | Improved |
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.
- T2ICount — T2ICount: Enhancing Cross-modal Understanding for Zero-Shot Counting (CVPR 2025) · better on 2 of 3 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)
- T2ICount: Enhancing Cross-modal Understanding for Zero-Shot Counting (CVPR 2025 · similarity 0.90)
- Single Domain Generalization for Few-Shot Counting via Universal Representation Matching (CVPR 2025 · similarity 0.88)
- Point Segment and Count: A Generalized Framework for Object Counting (CVPR 2024 · similarity 0.88)
- A Novel Unified Architecture for Low-Shot Counting by Detection and Segmentation (NeurIPS 2024 · similarity 0.88)
- DAVE - A Detect-and-Verify Paradigm for Low-Shot Counting (CVPR 2024 · similarity 0.88)
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.
