Grasp Any Region: Towards Precise, Contextual Pixel Understanding for Multimodal LLMs
This addresses the challenge of precise, contextual pixel understanding for multimodal AI systems, enabling active dialogue and compositional reasoning, though it is incremental in improving region-level methods.
The paper tackles the problem of fine-grained region-level understanding in multimodal LLMs by introducing Grasp Any Region (GAR), which uses RoI-aligned feature replay to incorporate global contexts and model interactions, resulting in state-of-the-art performance such as outperforming DAM-3B by +4.5 on DLC-Bench and surpassing InternVL3-78B on GAR-Bench-VQA.
While Multimodal Large Language Models (MLLMs) excel at holistic understanding, they struggle in capturing the dense world with complex scenes, requiring fine-grained analysis of intricate details and object inter-relationships. Region-level MLLMs have been a promising step. However, previous attempts are generally optimized to understand given regions in isolation, neglecting crucial global contexts. To address this, we introduce Grasp Any Region (GAR) for comprehen- sive region-level visual understanding. Empowered by an effective RoI-aligned feature replay technique, GAR supports (1) precise perception by leveraging necessary global contexts, and (2) modeling interactions between multiple prompts. Together, it then naturally achieves (3) advanced compositional reasoning to answer specific free-form questions about any region, shifting the paradigm from passive description to active dialogue. Moreover, we construct GAR-Bench, which not only provides a more accurate evaluation of single-region comprehension, but also, more importantly, measures interactions and complex reasoning across multiple regions. Extensive experiments have demonstrated that GAR-1B not only maintains the state-of-the-art captioning capabilities, e.g., outperforming DAM-3B +4.5 on DLC-Bench, but also excels at modeling relationships between multiple prompts with advanced comprehension capabilities, even surpassing InternVL3-78B on GAR-Bench-VQA. More importantly, our zero-shot GAR-8B even outperforms in-domain VideoRefer-7B on VideoRefer-BenchQ, indicating its strong capabilities can be easily transferred to videos.