Weakly Correlated Knowledge Integration for Few-shot Image Classification

Yang, Chun; Liu, Chang; Yin, Xu-Cheng

doi:10.1007/s11633-022-1320-9

Cited by 9 publications

(2 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Few-shot image classification [11][12][13][14] aims to classify new images with a few examples. Meta-learning [3,[19][20][21][22][23] is a common method to solve the few-shot problem.…”

Section: Few-shot Learning For Image Classificationmentioning

confidence: 99%

Multi‐similarity based hyperrelation network for few‐shot segmentation

Shi

Cui

Zhang

et al. 2022

IET Image Processing

View full text Add to dashboard Cite

Few-shot semantic segmentation aims at recognizing the object regions of unseen categories with only a few annotated examples as supervision. The key to few-shot segmentation is to establish a robust semantic relationship between the support and query images and to prevent overfitting. In this paper, an effective multi-similarity hyperrelation network (MSHNet) is proposed to tackle the few-shot semantic segmentation problem. In MSHNet, a new generative prototype similarity (GPS) is proposed, which, together with cosine similarity, establishes a strong semantic relationship between supported images and query images. In addition, a symmetric merging block (SMB) in MSHNet is proposed to efficiently merge multi-layer, multi-shot, multi-similarity features to generate hyperrelation features for semantic segmentation. Experimenting on two benchmark semantic segmentation datasets (Pascal−5 i and COCO−20 i ) shows that this method achieves a mean intersection-over-union score of 72.3% and 56.0%, respectively, which outperforms the state-of-the-art methods by 1.9% and 6.5%.

show abstract

“…Few-shot image classification [11][12][13][14] aims to classify new images with a few examples. Meta-learning [3,[19][20][21][22][23] is a common method to solve the few-shot problem.…”

Section: Few-shot Learning For Image Classificationmentioning

confidence: 99%

Multi‐similarity based hyperrelation network for few‐shot segmentation

Shi

Cui

Zhang

et al. 2022

IET Image Processing

View full text Add to dashboard Cite

show abstract

“…Its superiority could be primarily attributed to the accessibility of vast amount of supervised training data. Nevertheless, constrained by expertise and efforts, extensive data annotation could inevitably induce ambiguity and label noise, which might impose detrimental effects on model training (Wei et al 2022;Hu et al 2022;Yang, Liu, and Yin 2022). It is desirable to explore endowing modern learning systems with the power to deal with imperfect supervision.…”

Section: Introductionmentioning

confidence: 99%

Disentangled Partial Label Learning

Bao,

Rui,

Zhang

2024

AAAI

View full text Add to dashboard Cite

Partial label learning (PLL) induces a multi-class classifier from training examples each associated with a set of candidate labels, among which only one is valid. The formation of real-world data typically arises from heterogeneous entanglement of series latent explanatory factors, which are considered intrinsic properties for discriminating between different patterns. Though learning disentangled representation is expected to facilitate label disambiguation for partial-label (PL) examples, few existing works were dedicated to addressing this issue. In this paper, we make the first attempt towards disentangled PLL and propose a novel approach named TERIAL, which makes predictions according to derived disentangled representation of instances and label embeddings. The TERIAL approach formulates the PL examples as an undirected bipartite graph where instances are only connected with their candidate labels, and employs a tailored neighborhood routing mechanism to yield disentangled representation of nodes in the graph. Specifically, the proposed routing mechanism progressively infers the explanatory factors that contribute to the edge between adjacent nodes and augments the representation of the central node with factor-aware embedding information propagated from specific neighbors simultaneously via iteratively analyzing the promising subspace clusters formed by the node and its neighbors. The estimated labeling confidence matrix is also introduced to accommodate unreliable links owing to the inherent ambiguity of PLL. Moreover, we theoretically prove that the neighborhood routing mechanism will converge to the point estimate that maximizes the marginal likelihood of observed PL training examples. Comprehensive experiments over various datasets demonstrate that our approach outperforms the state-of-the-art counterparts.

show abstract

Region-adaptive Concept Aggregation for Few-shot Visual Recognition

Han

Zhan

et al. 2023

Mach. Intell. Res.

View full text Add to dashboard Cite

Weakly Correlated Knowledge Integration for Few-shot Image Classification

Cited by 9 publications

References 24 publications

Multi‐similarity based hyperrelation network for few‐shot segmentation

Multi‐similarity based hyperrelation network for few‐shot segmentation

Disentangled Partial Label Learning

Region-adaptive Concept Aggregation for Few-shot Visual Recognition

Contact Info

Product

Resources

About