Omni-RGPT: Unifying Image and Video Region-level Understanding via Token Marks

TL;DR

Omni-RGPT is a multimodal model that unifies image and video region understanding using Token Marks, enabling consistent region comprehension across spatial and temporal dimensions for various vision-language tasks.

Contribution

The paper introduces Token Mark tokens for region-level understanding and a large-scale dataset, advancing unified image and video comprehension in multimodal models.

Findings

Achieves state-of-the-art results on image and video reasoning benchmarks.

Demonstrates strong performance in captioning and referring expression tasks.

Supports robust video understanding without tracklets.

Abstract

We present Omni-RGPT, a multimodal large language model designed to facilitate region-level comprehension for both images and videos. To achieve consistent region representation across spatio-temporal dimensions, we introduce Token Mark, a set of tokens highlighting the target regions within the visual feature space. These tokens are directly embedded into spatial regions using region prompts (e.g., boxes or masks) and simultaneously incorporated into the text prompt to specify the target, establishing a direct connection between visual and text tokens. To further support robust video understanding without requiring tracklets, we introduce an auxiliary task that guides Token Mark by leveraging the consistency of the tokens, enabling stable region interpretation across the video. Additionally, we introduce a large-scale region-level video instruction dataset (RegVID-300k). Omni-RGPT achieves state-of-the-art results on image and video-based commonsense reasoning benchmarks while showing strong performance in captioning and referring expression comprehension tasks.