In defense of local descriptor-based few-shot object detection

被引:0
|
作者
Zhou, Shichao [1 ]
Li, Haoyan [1 ]
Wang, Zhuowei [1 ]
Zhang, Zekai [1 ]
机构
[1] Beijing Informat Sci & Technol Univ, Key Lab Informat & Commun Syst, Minist Informat Ind, Beijing, Peoples R China
基金
中国国家自然科学基金;
关键词
few-shot learning; local descriptors; contextual features; kernel method; visual similarity;
D O I
10.3389/fnins.2024.1349204
中图分类号
Q189 [神经科学];
学科分类号
071006 ;
摘要
State-of-the-art image object detection computational models require an intensive parameter fine-tuning stage (using deep convolution network, etc). with tens or hundreds of training examples. In contrast, human intelligence can robustly learn a new concept from just a few instances (i.e., few-shot detection). The distinctive perception mechanisms between these two families of systems enlighten us to revisit classical handcraft local descriptors (e.g., SIFT, HOG, etc.) as well as non-parametric visual models, which innately require no learning/training phase. Herein, we claim that the inferior performance of these local descriptors mainly results from a lack of global structure sense. To address this issue, we refine local descriptors with spatial contextual attention of neighbor affinities and then embed the local descriptors into discriminative subspace guided by Kernel-InfoNCE loss. Differing from conventional quantization of local descriptors in high-dimensional feature space or isometric dimension reduction, we actually seek a brain-inspired few-shot feature representation for the object manifold, which combines data-independent primitive representation and semantic context learning and thus helps with generalization. The obtained embeddings as pattern vectors/tensors permit us an accelerated but non-parametric visual similarity computation as the decision rule for final detection. Our approach to few-shot object detection is nearly learning-free, and experiments on remote sensing imageries (approximate 2-D affine space) confirm the efficacy of our model.
引用
收藏
页数:10
相关论文
共 50 条
  • [41] Recent Few-shot Object Detection Algorithms: A Survey with Performance Comparison
    Liu, Tianying
    Zhang, Lu
    Wang, Yang
    Guan, Jihong
    Fu, Yanwei
    Zhao, Jiajia
    Zhou, Shuigeng
    ACM TRANSACTIONS ON INTELLIGENT SYSTEMS AND TECHNOLOGY, 2023, 14 (04)
  • [42] Multi-spectral template matching based object detection in a few-shot learning manner
    Feng, Chen
    Cao, Zhiguo
    Xiao, Yang
    Fang, Zhiwen
    Zhou, Joey Tianyi
    INFORMATION SCIENCES, 2023, 624 : 20 - 36
  • [43] Few-Shot Emergency Siren Detection
    Cantarini, Michela
    Gabrielli, Leonardo
    Squartini, Stefano
    SENSORS, 2022, 22 (12)
  • [44] Robotic Object Perception Based on Multispectral Few-Shot Coupled Learning
    Xiong, Pengwen
    Tong, Xiaobao
    Liu, Peter X.
    Song, Aiguo
    Li, Zhijun
    IEEE TRANSACTIONS ON SYSTEMS MAN CYBERNETICS-SYSTEMS, 2023, 53 (10): : 6119 - 6131
  • [45] Few-shot Insider Threat Detection
    Yuan, Shuhan
    Zheng, Panpan
    Wu, Xintao
    Tong, Hanghang
    CIKM '20: PROCEEDINGS OF THE 29TH ACM INTERNATIONAL CONFERENCE ON INFORMATION & KNOWLEDGE MANAGEMENT, 2020, : 2289 - 2292
  • [46] Few-Shot Learning with Novelty Detection
    Bjerge, Kim
    Bodesheim, Paul
    Karstoft, Henrik
    DEEP LEARNING THEORY AND APPLICATIONS, PT I, DELTA 2024, 2024, 2171 : 340 - 363
  • [47] FEW-SHOT SOUND EVENT DETECTION
    Wang, Yu
    Salamon, Justin
    Bryan, Nicholas J.
    Bello, Juan Pablo
    2020 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH, AND SIGNAL PROCESSING, 2020, : 81 - 85
  • [48] A lightweight few-shot marine object detection network for unmanned surface vehicles
    Wang, Bo
    Jiang, Peng
    Gao, Jingxuan
    Huo, Wei
    Yang, Zhangqi
    Liao, Yulei
    OCEAN ENGINEERING, 2023, 277
  • [49] Few-Shot Class-Incremental Learning for Classification and Object Detection: A Survey
    Zhang, Jinghua
    Liu, Li
    Silven, Olli
    Pietikainen, Matti
    Hu, Dewen
    IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2025, 47 (04) : 2924 - 2945
  • [50] Multiscale Object Contrastive Learning-Derived Few-Shot Object Detection in VHR Imagery
    Chen, Jie
    Qin, Dengda
    Hou, Dongyang
    Zhang, Jun
    Deng, Min
    Sun, Geng
    IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 2022, 60