arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2310.03377cs.CV
收藏

ACT-Net: 手术视频中的锚点上下文动作检测

ACT-Net: Anchor-context Action Detection in Surgery Videos

University of Birmingham(伯明翰大学);Southern University of Science and Technology(南方科技大学);Research Institute of Trustworthy Autonomous Systems(可信自主系统研究院);National University of Singapore(新加坡国立大学);Third medical center of Chinese PLAGH(中国人民解放军总医院第三医学中心);Institute of High Performance Computing (IHPC), Agency for Science, Technology and Research (A*STAR)(新加坡科技研究局高性能计算研究所)

Luoying Hao, Yan Hu, Wenjun Lin, Qun Wang, Heng Li, Huazhu Fu, Jinming Duan, Jiang Liu

AI总结:

针对现有手术视频动作检测算法未利用手术流程规律导致准确率低、缺乏置信度评估的问题,提出含ACD和CCD模块的ACT-Net,实现高精度动作检测与置信度评估,mAP较基线提升4.0%。

摘要:

手术精细动作的识别与定位是开发上下文感知决策支持系统的核心组成部分。然而,现有大多数检测算法即使获得了动作位置,也无法输出高准确率的动作类别,原因在于它们未考虑整个视频中手术流程的规律性,这一局限性阻碍了其应用落地。此外,在临床应用中落地预测结果亟需传递模型置信度以获取信任,而这一点在手术动作预测领域尚未被探索。为精准检测每一时刻发生的细粒度动作,本文提出锚点上下文动作检测网络(ACT-Net),包含锚点上下文检测(ACD)模块和类别条件扩散(CCD)模块,用于解答三个问题:1)动作发生在何处;2)动作是什么;3)预测的置信度如何。具体而言,所提ACD模块在时空维度上高亮手术视频中与提取出的锚点存在交互的区域,基于锚点-上下文交互输出动作位置及其类别分布。考虑到视频中动作类别的完整分布,CCD模块采用以ACD估计器为条件的去噪扩散生成模型,进一步精准重构动作预测结果。此外,我们利用扩散模型输出的随机特性获取每个预测的模型置信度。本文方法取得了当前最优性能,在手术视频数据集上相比基线模型的平均精度均值(mAP)提升了4.0%。

英文摘要:

Recognition and localization of surgical detailed actions is an essential component of developing a context-aware decision support system. However, most existing detection algorithms fail to provide high-accuracy action classes even having their locations, as they do not consider the surgery procedure's regularity in the whole video. This limitation hinders their application. Moreover, implementing the predictions in clinical applications seriously needs to convey model confidence to earn entrustment, which is unexplored in surgical action prediction. In this paper, to accurately detect fine-grained actions that happen at every moment, we propose an anchor-context action detection network (ACTNet), including an anchor-context detection (ACD) module and a class conditional diffusion (CCD) module, to answer the following questions: 1) where the actions happen; 2) what actions are; 3) how confidence predictions are. Specifically, the proposed ACD module spatially and temporally highlights the regions interacting with the extracted anchor in surgery video, which outputs action location and its class distribution based on anchor-context interactions. Considering the full distribution of action classes in videos, the CCD module adopts a denoising diffusion-based generative model conditioned on our ACD estimator to further reconstruct accurately the action predictions. Moreover, we utilize the stochastic nature of the diffusion model outputs to access model confidence for each prediction. Our method reports the state-of-the-art performance, with improvements of 4.0% mAP against baseline on the surgical video dataset.

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。

↑