LARE: Low-Attention Region Encoding for Text-Image Retrieval
The paper proposes a framework called LARE to improve image retrieval in crowded scenes by modeling low-attention regions.
The paper proposes a novel framework that models low-attention regions in images, which is a new approach in the field of image retrieval.
Before reading this…
Applications
- →Image retrieval in crowded scenes
- →Object detection in complex images
To understand this paper, make sure you know these concepts first:
- Image processingfind papers →
- Computer visionfind papers →
Abstract
More Like ThisImage retrieval in crowded scenes is particularly challenging due to the salience bias of conventional visual encoders, which tend to focus on dominant objects while neglecting low-attention regions that are often crucial for fine-grained retrieval. We propose LARE (Low-Attention Region Encoding), a framework that explicitly models these overlooked regions. LARE adopts a dual-encoding strategy that encodes low-attention regions of an image and the full image in parallel, leading to more diverse and informative image embeddings. To evaluate image retrieval performance in challenging crowded scenes, we introduce Dense-Set, a challenging subset derived from COCO and Flickr30K. In this subset, images are re-captioned to provide richer descriptions of low-attention or previously overlooked regions. This dataset highlights the limitations of existing retrieval models and enables a more rigorous evaluation under densely crowded scene conditions. Experimental results demonstrate that the proposed framework improves retrieval performance by preserving subtle, non-dominant visual cues within the shared latent space.