用于答案定位的句子注意力模块
Sentence Attention Blocks for Answer Grounding
AI总结：
本文提出句子注意力模块，通过建模图像特征图与句子嵌入的依赖关系来校准通道特征，以简洁灵活的设计实现答案定位，并在多个数据集上达到最先进准确率。
摘要：
答案定位是为视觉问答任务定位相关视觉证据的任务。尽管针对该任务已提出多种多样的注意力方法，但它们存在以下三个问题：设计不允许使用预训练网络，因而无法受益于大规模数据预训练；定制设计未基于经过充分验证的先前设计，从而限制了网络的学习能力；或者设计过于复杂，导致难以重新实现或改进。在本文中，我们提出一种新颖的架构模块，称为句子注意力模块（Sentence Attention Block），以解决这些问题。所提出的模块通过显式建模图像特征图与句子嵌入之间的相互依赖关系，对通道维度的图像特征图进行重新校准。我们以可视化方式展示了该模块如何根据句子嵌入过滤掉不相关的特征图通道。我们从一种众所周知的注意力方法开始设计，并通过进行较小的修改，改进了结果，达到了最先进的准确率。我们方法的灵活性使其易于使用不同的预训练骨干网络，其简洁性使其易于理解和重新实现。我们在TextVQA-X、VQS、VQA-X和VizWiz-VQA-Grounding数据集上证明了我们方法的有效性。我们进行了多项消融研究，以展示我们设计选择的有效性。
英文摘要：
Answer grounding is the task of locating relevant visual evidence for the Visual Question Answering task. While a wide variety of attention methods have been introduced for this task, they suffer from the following three problems: designs that do not allow the usage of pre-trained networks and do not benefit from large data pre-training, custom designs that are not based on well-grounded previous designs, therefore limiting the learning power of the network, or complicated designs that make it challenging to re-implement or improve them. In this paper, we propose a novel architectural block, which we term Sentence Attention Block, to solve these problems. The proposed block re-calibrates channel-wise image feature-maps by explicitly modeling inter-dependencies between the image feature-maps and sentence embedding. We visually demonstrate how this block filters out irrelevant feature-maps channels based on sentence embedding. We start our design with a well-known attention method, and by making minor modifications, we improve the results to achieve state-of-the-art accuracy. The flexibility of our method makes it easy to use different pre-trained backbone networks, and its simplicity makes it easy to understand and be re-implemented. We demonstrate the effectiveness of our method on the TextVQA-X, VQS, VQA-X, and VizWiz-VQA-Grounding datasets. We perform multiple ablation studies to show the effectiveness of our design choices.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
