Zero-shot Referring Expression Comprehension via Structural Similarity Between Images and Captions
This work addresses the problem of fine-grained visual grounding without task-specific training, which is significant for applications in robotics and human-computer interaction, though it builds incrementally on existing vision-language alignment models.
The paper tackles zero-shot referring expression comprehension by localizing bounding boxes in images from textual prompts, achieving up to a 19.5% performance increase over the state-of-the-art zero-shot model on RefCOCO/+/g and comparable accuracy to fully supervised models on the Who's Waldo dataset.
Zero-shot referring expression comprehension aims at localizing bounding boxes in an image corresponding to provided textual prompts, which requires: (i) a fine-grained disentanglement of complex visual scene and textual context, and (ii) a capacity to understand relationships among disentangled entities. Unfortunately, existing large vision-language alignment (VLA) models, e.g., CLIP, struggle with both aspects so cannot be directly used for this task. To mitigate this gap, we leverage large foundation models to disentangle both images and texts into triplets in the format of (subject, predicate, object). After that, grounding is accomplished by calculating the structural similarity matrix between visual and textual triplets with a VLA model, and subsequently propagate it to an instance-level similarity matrix. Furthermore, to equip VLA models with the ability of relationship understanding, we design a triplet-matching objective to fine-tune the VLA models on a collection of curated dataset containing abundant entity relationships. Experiments demonstrate that our visual grounding performance increase of up to 19.5% over the SOTA zero-shot model on RefCOCO/+/g. On the more challenging Who's Waldo dataset, our zero-shot approach achieves comparable accuracy to the fully supervised model. Code is available at https://github.com/Show-han/Zeroshot_REC.