Vision Transformers have shown great promise recently for many vision tasks due to the insightful architecture design and attention mechanism. By revisiting the self-attention responses in Transformers, we empirically observe two interesting issues. First, Vision Transformers present a queryirrelevant behavior at deep layers, where the attention maps exhibit nearly consistent contexts in global scope, regardless of the query patch position (also head-irrelevant). Second, the attention maps are intrinsically sparse, few tokens dominate the attention weights; introducing the knowledge from ConvNets would largely smooth the attention and enhance the performance. Motivated by above observations, we generalize self-attention formulation to abstract a queryirrelevant global context directly and further integrate the global context into convolutions. The resulting model, a Fully Convolutional Vision Transformer (i.e., FCViT), purely consists of convolutional layers and firmly inherits the merits of both attention mechanism and convolutions, including dynamic property, weight sharing, and short- and long-range feature modeling, etc. Experimental results demonstrate the effectiveness of FCViT. With less than 14M parameters, our FCViT-S12 outperforms related work ResT-Lite by 3.7% top1 accuracy on ImageNet-1K. When scaling FCViT to larger models, we still perform better than previous state-of-the-art ConvNeXt with even fewer parameters. FCViT-based models also demonstrate promising transferability to downstream tasks, like object detection, instance segmentation, and semantic segmentation. Codes and models are made available at: https://github.com/ma-xu/FCViT.
translated by 谷歌翻译
Recently, webly supervised learning (WSL) has been studied to leverage numerous and accessible data from the Internet. Most existing methods focus on learning noise-robust models from web images while neglecting the performance drop caused by the differences between web domain and real-world domain. However, only by tackling the performance gap above can we fully exploit the practical value of web datasets. To this end, we propose a Few-shot guided Prototypical (FoPro) representation learning method, which only needs a few labeled examples from reality and can significantly improve the performance in the real-world domain. Specifically, we initialize each class center with few-shot real-world data as the ``realistic" prototype. Then, the intra-class distance between web instances and ``realistic" prototypes is narrowed by contrastive learning. Finally, we measure image-prototype distance with a learnable metric. Prototypes are polished by adjacent high-quality web images and involved in removing distant out-of-distribution samples. In experiments, FoPro is trained on web datasets with a few real-world examples guided and evaluated on real-world datasets. Our method achieves the state-of-the-art performance on three fine-grained datasets and two large-scale datasets. Compared with existing WSL methods under the same few-shot settings, FoPro still excels in real-world generalization. Code is available at https://github.com/yuleiqin/fopro.
translated by 谷歌翻译
As a powerful engine, vanilla convolution has promoted huge breakthroughs in various computer tasks. However, it often suffers from sample and content agnostic problems, which limits the representation capacities of the convolutional neural networks (CNNs). In this paper, we for the first time model the scene features as a combination of the local spatial-adaptive parts owned by the individual and the global shift-invariant parts shared to all individuals, and then propose a novel two-branch dual complementary dynamic convolution (DCDC) operator to flexibly deal with these two types of features. The DCDC operator overcomes the limitations of vanilla convolution and most existing dynamic convolutions who capture only spatial-adaptive features, and thus markedly boosts the representation capacities of CNNs. Experiments show that the DCDC operator based ResNets (DCDC-ResNets) significantly outperform vanilla ResNets and most state-of-the-art dynamic convolutional networks on image classification, as well as downstream tasks including object detection, instance and panoptic segmentation tasks, while with lower FLOPs and parameters.
translated by 谷歌翻译
事件参数提取(EAE)的目的是从文本中提取具有给定角色的参数,这些参数已在自然语言处理中得到广泛研究。以前的大多数作品在具有专用神经体系结构的特定EAE数据集中取得了良好的性能。鉴于,这些架构通常很难适应具有各种注释模式或格式的新数据集/方案。此外,他们依靠大规模标记的数据进行培训,由于大多数情况下的标签成本高,因此无法获得培训。在本文中,我们提出了一个具有变异信息瓶颈的多格式转移学习模型,该模型利用了信息,尤其是新数据集中EAE现有数据集中的常识。具体而言,我们引入了一个共享特定的及时框架,以从具有不同格式的数据集中学习格式共享和格式特定的知识。为了进一步吸收EAE的常识并消除无关的噪音,我们将变异信息瓶颈整合到我们的体系结构中以完善共享表示。我们在三个基准数据集上进行了广泛的实验,并在EAE上获得新的最先进的性能。
translated by 谷歌翻译
尽管在半监督语义细分领域的进度程度不同,但其最近的大部分成功都涉及笨拙的模型,并且尚未探索轻量级解决方案。我们发现,现有的知识蒸馏技术更多地关注标签数据中的像素级概念,该数据未能在未标记的数据中考虑更有用的线索。因此,我们提供了首次尝试通过新颖的多晶蒸馏(MGD)方案提供轻量级SSS模型,其中从三个方面捕获了多个跨性别:i)互补的教师结构; ii)标记为未标记的数据合作蒸馏; iii)分层和多层次损失设置。具体而言,MGD被配制为标记的未标记数据合作蒸馏方案,该方案有助于充分利用在半监督环境中必不可少的不同数据特征。图像水平的语义敏感损失,区域级别的内容感知损失和像素级的一致性损失是通过结构互补的教师来丰富层次蒸馏抽象的。 Pascal VOC2012和CityScapes的实验结果表明,在不同的分区协议下,MGD可以超越竞争方法。例如,在1/16的CityScapes分区协议下,RESNET-18和MOBILENET-V2主链的性能分别增长了11.5%和4.6%。尽管模型骨干的拖曳量被3.4-5.3倍(RESNET-18)和38.7-59.6X(MobileNetV2)压缩,但该模型旨在实现令人满意的分割结果。
translated by 谷歌翻译
对比学习在视频表示学习中表现出了巨大的潜力。但是,现有方法无法充分利用短期运动动态,这对于各种下游视频理解任务至关重要。在本文中,我们提出了运动敏感的对比度学习(MSCL),该学习将光学流捕获的运动信息注入RGB帧中,以增强功能学习。为了实现这一目标,除了剪辑级全球对比度学习外,我们还开发了局部运动对比度学习(LMCL),具有两种模式的框架级对比目标。此外,我们引入流动旋转增强(FRA),以生成额外的运动除件负面样品和运动差分采样(MDS)以准确筛选训练样品。对标准基准测试的广泛实验验证了该方法的有效性。以常用的3D RESNET-18为骨干,我们在UCF101上获得了91.5 \%的前1个精度,而在视频分类中进行了一些v2的v2,以及65.6 \%的top-1 top-1召回ucf1011对于视频检索,特别是改善了最新的。
translated by 谷歌翻译
视频异常检测(VAD)是计算机视觉中的重要主题。本文通过最新的自我监督学习进展的激励,通过解决直观而又具有挑战性的借口任务,即时空拼图拼图来解决VAD,该任务是一个多标签的精细粒度分类问题。我们的方法比现有作品具有几个优点:1)时空拼图难题是根据空间和时间维度分离的,分别捕获了高度歧视性的外观和运动特征; 2)完全排列用于提供涵盖各种难度水平的丰富拼图难题,从而使网络能够区分正常事件和异常事件之间的细微时空差异; 3)借口任务以端到端的方式解决,而无需依赖任何预训练的模型。我们的方法优于三个公共基准的最先进的方法。尤其是在上海校园中,其结果优于重建和基于预测的方法。
translated by 谷歌翻译
基于草图的3D形状检索(SBSR)是一项重要但艰巨的任务,近年来引起了越来越多的关注。现有方法在限制设置中解决了该问题,而无需适当模拟真实的应用程序方案。为了模仿现实的设置,在此曲目中,我们采用了不同级别的绘图技能的业余爱好者以及各种3D形状的大规模草图,不仅包括CAD型号,而且还可以从真实对象扫描的模型。我们定义了两个SBSR任务,并构建了两个基准,包括46,000多个CAD型号,1,700个现实型号和145,000个草图。四个团队参加了这一轨道,并为这两个任务提交了15次跑步,由7个常用指标评估。我们希望,基准,比较结果和开源评估法会在3D对象检索社区中促进未来的研究。
translated by 谷歌翻译
蛋白质是人类生命的重要组成部分,其结构对于功能和机制分析很重要。最近的工作表明了AI驱动方法对蛋白质结构预测的潜力。但是,新模型的开发受到数据集和基准测试培训程序的限制。据我们所知,现有的开源数据集远不足以满足现代蛋白质序列相关研究的需求。为了解决这个问题,我们介绍了具有高覆盖率和多样性的第一个百万级蛋白质结构预测数据集,称为PSP。该数据集由570K真实结构序列(10TB)和745K互补蒸馏序列(15TB)组成。此外,我们还提供了该数据集上SOTA蛋白结构预测模型的基准测试训练程序。我们通过参与客串比赛验证该数据集的实用程序进行培训,我们的模特赢得了第一名。我们希望我们的PSP数据集以及培训基准能够为AI驱动的蛋白质相关研究提供更广泛的AI/生物学研究人员社区。
translated by 谷歌翻译
最近,基于合成数据的实例分割已成为一种极其有利的优化范式,因为它利用模拟渲染和物理学来生成高质量的图像宣传对。在本文中,我们提出了一个并行预训练的变压器(PPT)框架,以完成基于合成数据的实例分割任务。具体而言,我们利用现成的预训练的视觉变压器来减轻自然数据和合成数据之间的差距,这有助于在下游合成数据场景中提供良好的概括,几乎没有样本。基于SWIN-B基的CBNET V2,基于SWINL的CBNET V2和SWIN-L基统一器用于并行特征学习,并且这三个模型的结果由像素级非最大最大抑制(NMS)算法融合来获得更强大的结果。实验结果表明,PPT在CVPR2022 AVA可访问性视觉和自主性挑战中排名第一,地图为65.155%。
translated by 谷歌翻译