Griffon v2: Advancing Multimodal Perception with High-Resolution Scaling and Visual-Language Co-Referring

TL;DR

Griffon v2 is a high-resolution multimodal model that enhances object perception and referring capabilities by scaling image resolution and integrating visual-language co-referring, surpassing existing models in complex scenarios.

Contribution

Introduces Griffon v2, a unified high-resolution multimodal model with a lightweight down-sampling projector and visual-language co-referring, enabling better perception and interaction.

Findings

Achieves state-of-the-art results on REC and phrase grounding.

Outperforms expert models in object detection and counting.

Effectively localizes objects with visual and textual references.

Abstract

Large Vision Language Models have achieved fine-grained object perception, but the limitation of image resolution remains a significant obstacle to surpassing the performance of task-specific experts in complex and dense scenarios. Such limitation further restricts the model's potential to achieve nuanced visual and language referring in domains such as GUI Agents, counting, \textit{etc}. To address this issue, we introduce a unified high-resolution generalist model, Griffon v2, enabling flexible object referring with visual and textual prompts. To efficiently scale up image resolution, we design a simple and lightweight down-sampling projector to overcome the input tokens constraint in Large Language Models. This design inherently preserves the complete contexts and fine details and significantly improves multimodal perception ability, especially for small objects. Building upon this, we further equip the model with visual-language co-referring capabilities through a plug-and-play visual tokenizer. It enables user-friendly interaction with flexible target images, free-form texts, and even coordinates. Experiments demonstrate that Griffon v2 can localize objects of interest with visual and textual referring, achieve state-of-the-art performance on REC and phrase grounding, and outperform expert models in object detection, object counting, and REG. Data and codes are released at https://github.com/jefferyZhan/Griffon.