Dragonfly: Multi-Resolution Zoom-In Encoding Enhances Vision-Language Models

• URL: http://arxiv.org/abs/2406.00977v2
• Date: Mon, 14 Oct 2024 23:15:24 GMT
• Title: Dragonfly: Multi-Resolution Zoom-In Encoding Enhances Vision-Language Models
• Authors: Rahul Thapa, Kezhen Chen, Ian Covert, Rahul Chalamala, Ben Athiwaratkun, Shuaiwen Leon Song, James Zou,
• Abstract summary: Vision transformers (ViTs) struggle to capture fine-grained details from less prominent objects, charts, and embedded text. We extend recent high-resolution and multi-crop techniques by not only preserving the native resolution, but zooming in beyond it. This enhancement allows our model to better capture fine-grained details, overcoming the limitations of current ViTs.
• Score: 26.322856874796702
• License:
• Abstract: Recent advances in vision-language models (VLMs) have demonstrated the advantages of processing images at higher resolutions and utilizing multi-crop features to preserve native resolution details. However, despite these improvements, existing vision transformers (ViTs) still struggle to capture fine-grained details from less prominent objects, charts, and embedded text, limiting their effectiveness in certain tasks. In this paper, we extend recent high-resolution and multi-crop techniques by not only preserving the native resolution, but zooming in beyond it and extracting features from a large number of image sub-crops. This enhancement allows our model to better capture fine-grained details, overcoming the limitations of current ViTs. To manage the increased token count and computational complexity, we demonstrate that a simple mean-pooling aggregation over tokens is effective. Our model, Dragonfly, achieves competitive performance on general-domain tasks such as ScienceQA and AI2D, and excels in tasks requiring fine-grained image understanding, including TextVQA and ChartQA. Among models in the 7-8B parameter range, Dragonfly consistently ranks at the top across ten general-domain benchmarks, achieving the highest or second-highest scores in most cases, outperforming models that are significantly larger or trained on larger datasets. Our biomedical model, Dragonfly-Med, sets new benchmarks on several medical tasks, achieving 91.6% accuracy on SLAKE (compared to 84.8% for Med-Gemini), a 67.1% token F1 score on Path-VQA (compared to 62.7% for Med-PaLM M), and state-of-the-art results across the majority of image captioning tasks. Overall, our work highlights the persistent challenge of engineering visual representations with fixed-resolution ViTs, and proposes a simple yet effective solution to address this issue and boost performance in both general and specialized domains.

Related papers

• Griffon-G: Bridging Vision-Language and Vision-Centric Tasks via Large Multimodal Models [27.45225442048711]
  We introduce CCMD-8M, which overcomes the data barriers of unifying vision-centric and vision-language tasks. We also present Griffon-G, a general large multimodal model that addresses both vision-centric and vision-language tasks within a single end-to-end paradigm.
  arXiv  Detail & Related papers  (2024-10-21T16:30:29Z)
• VEGA: Learning Interleaved Image-Text Comprehension in Vision-Language Large Models [76.94378391979228]
  We introduce a new, more demanding task known as Interleaved Image-Text (IITC) This task challenges models to discern and disregard superfluous elements in both images and text to accurately answer questions. In support of this task, we further craft a new VEGA dataset, tailored for the IITC task on scientific content, and devised a subtask, Image-Text Association (ITA)
  arXiv  Detail & Related papers  (2024-06-14T17:59:40Z)
• Evaluating the Efficacy of Prompt-Engineered Large Multimodal Models Versus Fine-Tuned Vision Transformers in Image-Based Security Applications [2.8161155726745237]
  Large Multimodal Models (LMMs) are designed to interpret and analyze complex data by integrating multiple modalities such as text and images. This paper investigates the applicability and effectiveness of prompt-engineered LMMs that process both images and text, compared to fine-tuned Vision Transformer (ViT) models. For the visually non-evident task, the results highlight a significant divergence in performance, with ViT models achieving F1-scores of 97.11% in predicting 25 malware classes and 97.61% in predicting 5 malware families.
  arXiv  Detail & Related papers  (2024-03-26T15:20:49Z)
• Pushing Boundaries: Exploring Zero Shot Object Classification with Large Multimodal Models [0.09264362806173355]
  Large Language and Vision Assistant models (LLVAs) engage users in rich conversational experiences intertwined with image-based queries. This paper takes a unique perspective on LMMs, exploring their efficacy in performing image classification tasks using tailored prompts. Our study includes a benchmarking analysis across four diverse datasets: MNIST, Cats Vs. Dogs, Hymnoptera (Ants Vs. Bees), and an unconventional dataset comprising Pox Vs. Non-Pox skin images.
  arXiv  Detail & Related papers  (2023-12-30T03:19:54Z)
• Jack of All Tasks, Master of Many: Designing General-purpose Coarse-to-Fine Vision-Language Model [83.85856356798531]
  VistaLLM is a visual system that addresses coarse- and fine-grained vision-language tasks. It employs a gradient-aware adaptive sampling technique to represent binary segmentation masks as sequences. We also introduce a novel task, AttCoSeg, which boosts the model's reasoning and grounding capability over multiple input images.
  arXiv  Detail & Related papers  (2023-12-19T18:53:01Z)
• An Efficient General-Purpose Modular Vision Model via Multi-Task Heterogeneous Training [79.78201886156513]
  We present a model that can perform multiple vision tasks and can be adapted to other downstream tasks efficiently. Our approach achieves comparable results to single-task state-of-the-art models and demonstrates strong generalization on downstream tasks.
  arXiv  Detail & Related papers  (2023-06-29T17:59:57Z)
• AMIGO: Sparse Multi-Modal Graph Transformer with Shared-Context Processing for Representation Learning of Giga-pixel Images [53.29794593104923]
  We present a novel concept of shared-context processing for whole slide histopathology images. AMIGO uses the celluar graph within the tissue to provide a single representation for a patient. We show that our model is strongly robust to missing information to an extent that it can achieve the same performance with as low as 20% of the data.
  arXiv  Detail & Related papers  (2023-03-01T23:37:45Z)
• Advancing Plain Vision Transformer Towards Remote Sensing Foundation Model [97.9548609175831]
  We resort to plain vision transformers with about 100 million parameters and make the first attempt to propose large vision models customized for remote sensing tasks. Specifically, to handle the large image size and objects of various orientations in RS images, we propose a new rotated varied-size window attention. Experiments on detection tasks demonstrate the superiority of our model over all state-of-the-art models, achieving 81.16% mAP on the DOTA-V1.0 dataset.
  arXiv  Detail & Related papers  (2022-08-08T09:08:40Z)
• MaxViT: Multi-Axis Vision Transformer [19.192826213493838]
  We introduce an efficient and scalable attention model we call multi-axis attention. We present a new architectural element by effectively blending our proposed attention model with convolutions. We demonstrate the effectiveness of our model on a broad spectrum of vision tasks.
  arXiv  Detail & Related papers  (2022-04-04T17:59:44Z)
• VOLO: Vision Outlooker for Visual Recognition [148.12522298731807]
  Vision transformers (ViTs) have shown great potential of self-attention based models in ImageNet classification. We introduce a novel outlook attention and present a simple and general architecture, termed Vision Outlooker (VOLO) Unlike self-attention that focuses on global dependency modeling at a coarse level, the outlook attention efficiently encodes finer-level features and contexts into tokens. Experiments show that our VOLO achieves 87.1% top-1 accuracy on ImageNet-1K classification, which is the first model exceeding 87% accuracy on this competitive benchmark.
  arXiv  Detail & Related papers  (2021-06-24T15:46:54Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.