OMG-LLaVA: Bridging Image-level, Object-level, Pixel-level Reasoning and Understanding

TL;DR

OMG-LLaVA is a unified framework that combines pixel-level image understanding with reasoning and language capabilities, enabling flexible multimodal interactions and surpassing specialized methods on multiple benchmarks.

Contribution

It introduces a novel end-to-end model integrating universal segmentation with large language models for comprehensive visual reasoning and understanding.

Findings

Achieves image, object, and pixel-level reasoning in a single model.

Surpasses performance of specialized methods on multiple benchmarks.

Supports flexible user interaction with visual and text prompts.

Abstract

Current universal segmentation methods demonstrate strong capabilities in pixel-level image and video understanding. However, they lack reasoning abilities and cannot be controlled via text instructions. In contrast, large vision-language multimodal models exhibit powerful vision-based conversation and reasoning capabilities but lack pixel-level understanding and have difficulty accepting visual prompts for flexible user interaction. This paper proposes OMG-LLaVA, a new and elegant framework combining powerful pixel-level vision understanding with reasoning abilities. It can accept various visual and text prompts for flexible user interaction. Specifically, we use a universal segmentation method as the visual encoder, integrating image information, perception priors, and visual prompts into visual tokens provided to the LLM. The LLM is responsible for understanding the user's text instructions and providing text responses and pixel-level segmentation results based on the visual information. We propose perception prior embedding to better integrate perception priors with image features. OMG-LLaVA achieves image-level, object-level, and pixel-level reasoning and understanding in a single model, matching or surpassing the performance of specialized methods on multiple benchmarks. Rather than using LLM to connect each specialist, our work aims at end-to-end training on one encoder, one decoder, and one LLM. The code and model have been released for further research.