IntRec: Intent-based Retrieval with Contrastive Refinement

Retrieving user-specified objects from complex scenes remains a challenging task, especially when queries are ambiguous or involve multiple similar objects. Existing open-vocabulary detectors operate in a one-shot manner, lacking the ability to refine predictions based on user feedback. To address this, we propose IntRec, an interactive object retrieval framework that refines predictions based on user feedback. At its core is an Intent State (IS) that maintains dual memory sets for positive anchors (confirmed cues) and negative constraints (rejected hypotheses). A contrastive alignment function ranks candidate objects by maximizing similarity to positive cues while penalizing rejected ones, enabling fine-grained disambiguation in cluttered scenes. Our interactive framework provides substantial improvements in retrieval accuracy without additionalsupervision. OnLVIS, IntRecachieves35. 4AP, outperforming OVMR, CoDet, and CAKE by +2. 3, +3. 7, and +0. 5, respectively. On the challenging LVIS-Ambiguous benchmark, it improves performanceby+7. 9APoveritsone-shotbaselineafterasinglecorrective feedback, with less than 30 ms of added latency per interaction.

Authors

• Pourya Shamsolmoali Dept. Computer Science, University of York
• Masoumeh Zareapoor Dept. Computer Science, Shanghai Jiao Tong University
• Eric Granger LIVIA, Dept. of Systems Engineering, École de technologie supérieure
• Yue Lu School of Communication and Electronic Eng. , East China Normal University

Keywords

• Interactive learning
• Contrastive alignment
• Object detection