Generalized-Scale Object Counting with Gradual Query Aggregation
Jer Pelhan, Alan Lukežič, Matej Kristan · 2025
Read Generalized-Scale Object Counting with Gradual Query Aggregation by Jer Pelhan, Alan Lukežič, Matej Kristan (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2511-08048.txt.
Description
Jer Pelhan, Alan Lukezic, and Matej Kristan present GECO2, a few shot counter that finds objects given only a few example images. Counting across very different object sizes is hard, and pipelines often upscale and tile images to catch tiny items in crowds. Those fixes raise memory cost and still struggle when sizes vary widely within one picture.
GECO2 uses a dense query representation that gradually gathers exemplar features across scales. The result is a high resolution query set able to detect both large and small objects in one pass. It improves counting and detection accuracy by about ten percent over leading few shot counters while running three times faster.
Catalog record
| CREATOR | Jer Pelhan, Alan Lukežič, Matej Kristan |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 34 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2511-08048