Ming Yan, Ji Zhang, Liang He, Jiabo Ye, Xiaoshan Yang, Junfeng Tian, Xin Lin, Xuwu Wang
We have not lifted any functions out of this paper's repositories yet, so there is nothing we have run. If it links a repository, it is listed below.
Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.
Visual grounding focuses on establishing fine-grained alignment between vision and natural language, which has essential applications in multimodal reasoning systems. Existing methods use pre-trained query-agnostic visual backbones to extract visual feature maps independently without considering the query information. We argue that the visual features extracted from the visual backbones and the features really needed for multimodal reasoning are inconsistent. One reason is that there are differences between pretraining tasks and visual grounding. Moreover, since the backbones are query-agnostic, it is difficult to completely avoid the inconsistency issue by training the visual backbone end-to-end in the visual grounding framework. In this paper, we propose a Query-modulated Refinement Network (QRNet) to address the inconsistent issue by adjusting intermediate features in the visual backbone with a novel Queryaware Dynamic Attention (QD-ATT) mechanism and queryaware multiscale fusion. The QD-ATT can dynamically compute query-dependent visual attention at the spatial and channel levels of the feature maps produced by the visual backbone. We apply the QRNet to an end-to-end visual grounding framework. Extensive experiments show that the proposed method outperforms state-of-the-art methods on five widely used datasets. Our code is available at https: //github.com/LukeForeverYoung/QRNet.
The same record, over MCP at https://syntology.ai/mcp:
get_harvested_code_for_paper("2203.15442")
get_code_for_paper("2203.15442")
have("2203.15442")
Connect an agent — have() is free.