LLaVA-Grounding: Grounded Visual Chat with Large Multimodal Models
This addresses the need for improved visual chat systems that can accurately ground objects in images, though it is incremental as it builds on existing multimodal models.
The paper tackles the problem of large multimodal models lacking combined grounding and chat capabilities by creating a grounded visual chat dataset and benchmark, resulting in a model that outperforms others on the new benchmark and achieves competitive performance on classic grounding benchmarks.
With the recent significant advancements in large multi-modal models (LMMs), the importance of their grounding capability in visual chat is increasingly recognized. Despite recent efforts to enable LMMs to support grounding, their capabilities for grounding and chat are usually separate, and their chat performance drops dramatically when asked to ground. The problem is the lack of a dataset for grounded visual chat (GVC). Existing grounding datasets only contain short captions. To address this issue, we have created GVC data that allows for the combination of grounding and chat capabilities. To better evaluate the GVC capabilities, we have introduced a benchmark called Grounding-Bench. Additionally, we have proposed a model design that can support GVC and various types of visual prompts by connecting segmentation models with language models. Experimental results demonstrate that our model outperforms other LMMs on Grounding-Bench. Furthermore, our model achieves competitive performance on classic grounding benchmarks like RefCOCO/+/g and Flickr30K Entities. Our code will be released at https://github.com/UX-Decoder/LLaVA-Grounding .