RECOUNT: Reference-guided Counting with Synthetic Visual Exemplars

Researchers introduced RECOUNT, a plug-and-play framework for image-guided zero-shot object counting that uses a single off-scene reference image and a diffusion model to generate a diverse exemplar gallery, improving counting accuracy by 55% on LookAlikes and 21% on PairTally compared to the strongest prior zero-shot counter.

arXiv:2608.20621v1 Announce Type: new Abstract: Text-guided zero-shot object counters excel at spatial localization but categorize poorly on novel or fine-grained classes: natural language is too coarse to fully specify visual identity, so they fail to separate visually similar distractors. Few-shot counters sidestep this with visual exemplars, but require manual annotations on every image. To resolve this dilemma, we introduce RECOUNT, a plug-and-play framework for image-guided zero-shot counting. Rather than specify a category with a text prompt, our key insight is to specify it visually, from a single off-scene reference image. However, we find that a lone reference image provides narrow coverage of a category's appearance and is unreliable across diverse scenes. We therefore repurpose a diffusion model as an automated contrastive data engine that expands the reference into a diverse exemplar gallery, supplying the discriminative detail that text cannot. RECOUNT preserves the class-agnostic proposals of any frozen counter and offloads categorization to a separate visual module a frozen backbone with a lightweight head trained on this synthetic data that matches each proposal against the target and distractor galleries. Applied to a frozen counter, RECOUNT attains the best zero-shot accuracy on both benchmarks, cutting counting error MAE by 55% on LookAlikes and 21% on PairTally relative to the strongest prior zero-shot counter.