Few-shot classification (FSC) is one of the most concerned hot issues in recent years. The general setting consists of two phases: (1) Pre-train a feature extraction model (FEM) with base data (has large amounts of labeled samples). (2) Use the FEM to extract the features of novel data (with few labeled samples and totally different categories from base data), then classify them with the to-be-designed classifier. The adaptability of pre-trained FEM to novel data determines the accuracy of novel features, thereby affecting the final classification performances. To this end, how to appraise the pre-trained FEM is the most crucial focus in the FSC community. It sounds like traditional Class Activate Mapping (CAM) based methods can achieve this by overlaying weighted feature maps. However, due to the particularity of FSC (e.g., there is no backpropagation when using the pre-trained FEM to extract novel features), we cannot activate the feature map with the novel classes. To address this challenge, we propose a simple, flexible method, dubbed as Class-Irrelevant Mapping (CIM). Specifically, first, we introduce dictionary learning theory and view the channels of the feature map as the bases in a dictionary. Then we utilize the feature map to fit the feature vector of an image to achieve the corresponding channel weights. Finally, we overlap the weighted feature map for visualization to appraise the ability of pre-trained FEM on novel data. For fair use of CIM in evaluating different models, we propose a new measurement index, called Feature Localization Accuracy (FLA). In experiments, we first compare our CIM with CAM in regular tasks and achieve outstanding performances. Next, we use our CIM to appraise several classical FSC frameworks without considering the classification results and discuss them.
Slot Attention-based Feature Filtering for Few-Shot Learning
AIBA, Departament de Matemàtiques & Informàtica Universitat de Barcelona(AIBA,数学与信息学院巴塞罗那大学);Departamento de Ingeniería de Sistemas y Computación Universidad Católica del Norte(系统与计算系卡罗林大学);Institute of Neuroscience, Universitat de Barcelona(神经科学研究所巴塞罗那大学)
Few-Shot Learning with Visual Distribution Calibration and Cross-Modal Distribution Alignment
Beihang University(北京航空航天大学);Huawei Noah’s Ark Lab(华为诺亚方舟实验室);Tokyo Institute of Technology(东京工业大学);University of Science and Technology of China(中国科学技术大学);Zhongguancun Laboratory(中关村实验室)
Object-level Correlation for Few-Shot Segmentation
School of Computer Science and Engineering, Southeast University(东南大学计算机科学与工程学院);Samsung Electronics (China) R&D Centre(三星电子(中国)研发中心);Shanghai AI Laboratory(上海人工智能实验室);Institute for Infocomm Research (I2R), A*STAR Singapore(资讯通信研究院(I2R),新加坡A*STAR);Center for Machine Vision and Signal Analysis (CMVS), University of Oulu(图像与信号分析中心(CMVS),奥卢大学)
Self-Paced Uncertainty Estimation for One-shot Person Re-Identification
Beijing Institute of Technology(北京理工大学)
在构建从$X$到$Y$的预测模型时,一次性人员重新识别场景面临两种不确定性。第一个是模型不确定性,由于缺乏训练数据,它会捕获DNN中参数的噪声。第二种是数据不确定性,它可以分为两个子类型:一种是图像噪声,其中严重的遮挡和复杂的背景包含与身份无关的信息;另一种是标签噪声,其中错误标记影响视觉外观学习。针对这些问题,本文提出了一种新的用于one-shot Person Re-ID的自步调不确定性估计网络(SPUE-Net),通过引入自步调采样策略,该方法可以迭代地估计未标记样本的伪标签,从而逐渐扩展标记样本,消除模型的不确定性,而无需额外的监督。我们将伪标签样本分成两个子集,使训练样本的使用更加合理和有效。此外,本文采用局部不确定性估计与确定性估计相结合的协同学习方法,实现了更好的隐藏空间特征挖掘,提高了伪标记样本选取的精度,降低了数据的不确定性。基于视频和图像的数据集上的广泛的比较评估实验表明,SPUE-Net具有显着的优势,比国家的最先进的方法。
Denoise-I2W: Mapping Images to Denoising Words for Accurate Zero-Shot Composed Image Retrieval
Institute of Information Engineering, Chinese Academy of Sciences(中国科学院信息工程研究所);School of Cyber Security, University of Chinese Academy of Sciences(中国科学院大学网络空间安全学院);School of Information Engineering, Minzu University of China(中央民族大学信息工程学院);School of Cyberspace Science and Technology, Beijing Institute of Technology(北京理工大学网络空间科学与技术学院)
Stable Attribute Group Editing for Reliable Few-shot Image Generation
University of Chinese Academy of Sciences(中国科学院大学);Institute of Computing Technology, Chinese Academy of Sciences(中国科学院计算技术研究所);Peng Cheng Laboratory(鹏城实验室);Huawei Cloud EI Innovation Lab(华为云EI创新实验室)
Unleashing In-context Learning of Autoregressive Models for Few-shot Image Manipulation
GenAI, Meta(Meta GenAI部门);University of Wisconsin–Madison(威斯康星大学麦迪逊分校);University of Illinois Urbana-Champaign(伊利诺伊大学厄巴纳-香槟分校);Sungkyunkwan University(成均馆大学);Georgia Institute of Technology(佐治亚理工学院)
Mask-guided Vision Transformer (MG-ViT) for Few-Shot Learning
University of Electronic Science and Technology of China(电子科技大学);University of Georgia(佐治亚大学);University of Texas at Arlington(德克萨斯大学阿灵顿分校);Northwestern Polytechnical University(西北工业大学)
Prompt-and-Transfer: Dynamic Class-aware Enhancement for Few-shot Segmentation
Aerospace Information Research Institute, Chinese Academy of Sciences(中国科学院空天信息创新研究院);School of Electronic, Electrical and Communication Engineering, University of Chinese Academy of Sciences(中国科学院大学电子电气与通信工程学院);University of Chinese Academy of Sciences(中国科学院大学);Tsinghua University(清华大学)
D^3-Talker: Dual-Branch Decoupled Deformation Fields for Few-Shot 3D Talking Head Synthesis
School of Computer Science(计算机科学学院);Software Engineering, Shenzhen University(软件工程,深圳大学);Computer Vision Institute, School of Artificial Intelligence, Shenzhen University, Shenzhen, China(计算机视觉研究院,人工智能学院,深圳大学,深圳,中国);Guangdong Provincial Key Laboratory of Intelligent Information Processing(广东省智能信息处理重点实验室);HBUG Lab, University of Exeter(HBUG实验室,埃克塞特大学)
Not Just Learning from Others but Relying on Yourself: A New Perspective on Few-Shot Segmentation in Remote Sensing
Aerospace Information Research Institute, Chinese Academy of Sciences(中国科学院空天信息创新研究院);School of Electronic, Electrical and Communication Engineering, University of Chinese Academy of Sciences(中国科学院大学电子电气与通信工程学院);University of Chinese Academy of Sciences(中国科学院大学)