Papers › A Novel Unified Architecture for Low-Shot Counting by Detection and Segmentation

A Novel Unified Architecture for Low-Shot Counting by Detection and Segmentation

27 Sep 2024arXiv:2409.18686archive 2025-07-28

Jer Pelhan, Alan Lukežič, Vitjan Zavrtanik, Matej Kristan

Low-shot object counters estimate the number of objects in an image using few or no annotated exemplars. Objects are localized by matching them to prototypes, which are constructed by unsupervised image-wide object appearance aggregation. Due to potentially diverse object appearances, the existing approaches often lead to overgeneralization and false positive detections. Furthermore, the best-performing methods train object localization by a surrogate loss, that predicts a unit Gaussian at each object center. This loss is sensitive to annotation error, hyperparameters and does not directly optimize the detection task, leading to suboptimal counts. We introduce GeCo, a novel low-shot counter that achieves accurate object detection, segmentation, and count estimation in a unified architecture. GeCo robustly generalizes the prototypes across objects appearances through a novel dense object query formulation. In addition, a novel counting loss is proposed, that directly optimizes the detection task and avoids the issues of the standard surrogate loss. GeCo surpasses the leading few-shot detection-based counters by ∼25\% in the total count MAE, achieves superior detection accuracy and sets a new solid state-of-the-art result across all low-shot counting setups.

PaperPDFCode

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

jerpelhan/GeCo officialmentioned on GitHubpytorch report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Exemplar-Free CountingFew-shot Object Counting and DetectionObjectObject CountingObject DetectionObject Localizationobject-detection

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Exemplar-Free Counting FSC147 GeCo MAE(test) 13.30 #2 of 9 Archive leaderboard report
Exemplar-Free Counting FSC147 GeCo MAE(val) 14.81 #2 of 9 Archive leaderboard report
Exemplar-Free Counting FSC147 GeCo RMSE(test) 108.72 #2 of 9 Archive leaderboard report
Exemplar-Free Counting FSC147 GeCo RMSE(val) 64.95 #2 of 9 Archive leaderboard report
Few-shot Object Counting and Detection FSC147 GeCo AP(test) 43.42 #1 of 4 Archive leaderboard report
Few-shot Object Counting and Detection FSC147 GeCo AP50(test) 75.06 #1 of 4 Archive leaderboard report
Few-shot Object Counting and Detection FSC147 GeCo MAE(test) 7.91 #1 of 4 Archive leaderboard report
Few-shot Object Counting and Detection FSC147 GeCo RMSE(test) 54.28 #1 of 4 Archive leaderboard report
Object Counting FSC147 GeCo MAE(test) 7.91 #2 of 19 Archive leaderboard report
Object Counting FSC147 GeCo MAE(val) 9.52 #2 of 19 Archive leaderboard report
Object Counting FSC147 GeCo RMSE(test) 54.28 #2 of 19 Archive leaderboard report
Object Counting FSC147 GeCo RMSE(val) 43.00 #2 of 19 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

GECOMAE

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections