Zero-shot action recognition (ZSAR) is a practical and challenging issue, which compensates for the shortcomings of existing action recognition by being able to recognize those action classes that don't have visual representation during training. However, existing zero-shot action recognition doesn't...
因为是Zero-Shot Action Recognition(ZSAR),和ZSL一样,涉及到三个concept:video、attribute、label/action。两种常见的方法如上图的(a)、(b)所示,一是通过对attributes和video的embedding,然后通过knowledge transfer,来对unseen类的action进行识别分类(action-attribute)。另一种是用语义表达semantic representations(如常...
2019_Out-of-Distribution Detection for Generalized Zero-Shot Action Recognition 2019_CVPR_CEWGAN-OD_特征熵分出seen和unseen_GAN生成unseen_重构损失_生成特征的类别匹配损失 对于seen类上进行GAN训练, 从而合成unseen的动作特征. 首次引入 out-of-distribution detector 来确定是属于seen还是unseen的动作类别. seen...
Generalizedzero-shotactionrecognition中对可见和不可见动作类别进行单独处理来解决该问题。 本文引入了out-of-distribution检测器,该检测器确定视频特征...的基于out-of-distribution检测器的Generalizedzero-shotactionrecognition框架: OD检测器可以减少标准GZSL框架中存在的可见动作类别的固有 ...
Zero-shot action recognition requires a strong ability to generalize from pre-training and seen classes to novel unseen classes. Similarly, continual learning aims to develop models that can generalize effectively and learn new tasks without forgetting the ones previously learned. The generalization goals...
Zero-shot action recognition (ZSAR) aims to learn an alignment model between videos and class descriptions of seen actions that is transferable to unseen actions. The text queries (class descriptions) used in existing ZSAR works, however, are often short action names that fail to capture the ...
In this paper, we address zero-shot recognition in contemporary video action recognition tasks, using semantic word vector space as the common space to embed videos and category labels. This is more challenging because the mapping between the semantic space and space-time features of videos ...
Recently, zero-shot action recognition (ZSAR) has emerged with the explosive growth of action categories. In this paper, we explore ZSAR from a novel perspective by adopting the Error-Correcting Output Codes (dubbed ZSECOC). Our ZSECOC equips the conventional ECOC with the additional capability...
零次学习(zero-shot learning) 众所周知,深度学习的崛起依赖于大量的训练样本;监督式学习已经在各项任务上取得了极好的效果。 但有一点和我们人的“智能”不一样的是,一个两岁稚子第一次进动物园,看到老虎时,由于在家中见过猫,根据其父的描述“和猫很像、但比猫更大、丑恶更丑的极有可能是老虎”即可轻易判断...
实验分了两个方面,一是zer-shot visual recognition,在AWA和CUB数据上进行的;另一个是human action recognition,在UCF101和HMDB51上进行的。文章还给了很多数据集的细节和visual、semantic的编码表达细节,以及超参数的设置。这里就直接贴结果了: 总的来说,这个效果在2017年还是不错的,而且方法比较新颖,也有进步的空间...