Towards Cross-Granularity Few-Shot Learning: Coarse-to-Fine Pseudo-Labeling with Visual-Semantic Meta-Embedding

Yang, Jinhai; Yang, Hua; Lin, Chen

doi:10.1145/3474085.3475200

Cited by 15 publications

(5 citation statements)

References 53 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Based on the hierarchical relation between labels of different granularity, their method assigns fine-grained pseudo-labels to unlabeled images, which are used for model training. Another few-shot method that relies on pseudo-labeling for FGVC, was proposed by Yang et al [13]. In particular, they perform clustering of coarse-grained-class instances into pseudo-fine-grained classes and learn fine-grained features, which are then used for the actual task of FGVC.…”

Section: Fine-grained Classification Without Labelled Datamentioning

confidence: 99%

“…Despite the strong assumption of literal occurrence, that work concluded that indexing at this level of detail is useful for the precise retrieval of relevant documents. In this direction, the indexing policies for librarians of the CISMeF catalogue 13 for French medical resources were updated to support indexing at the level of MeSH concepts.…”

Section: Towards Fine-grained Semantic Indexing Of Biomedical Literaturementioning

confidence: 99%

“…https://www.nlm.nih.gov/mesh/qualifiers_scopenotes.html 12 https://www.nlm.nih.gov/mesh/intro_record_types.html 13. http://www.chu-rouen.fr/cismef/…”

mentioning

confidence: 99%

See 2 more Smart Citations

Large-scale fine-grained semantic indexing of biomedical literature based on weakly-supervised deep learning

Nentidis¹,

Chatzopoulos²,

Krithara³

et al. 2023

Preprint

View full text Add to dashboard Cite

Semantic indexing of biomedical literature is usually done at the level of MeSH descriptors, representing topics of interest for the biomedical community. Several related but distinct biomedical concepts are often grouped together in a single coarse-grained descriptor and are treated as a single topic for semantic indexing. This study proposes a new method for the automated refinement of subject annotations at the level of concepts, investigating deep learning approaches. Lacking labelled data for this task, our method relies on weak supervision based on concept occurrence in the abstract of an article. The proposed approach is evaluated on an extended large-scale retrospective scenario, taking advantage of concepts that eventually become MeSH descriptors, for which annotations become available in MEDLINE/PubMed. The results suggest that concept occurrence is a strong heuristic for automated subject annotation refinement and can be further enhanced when combined with dictionary-based heuristics. In addition, such heuristics can be useful as weak supervision for developing deep learning models that can achieve further improvement in some cases.

show abstract

Section: Fine-grained Classification Without Labelled Datamentioning

confidence: 99%

Section: Towards Fine-grained Semantic Indexing Of Biomedical Literaturementioning

confidence: 99%

See 1 more Smart Citation

Large-scale fine-grained semantic indexing of biomedical literature based on weakly-supervised deep learning

Nentidis¹,

Chatzopoulos²,

Krithara³

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…Judging from the fine-class stage (Fig. 1 middle to right), if we combine a pre-training set and the support set as a holistic training set, then the few-shot fine-grained recognition using a model pre-trained on coarse samples are similar to the weakly-supervised learning and specifically learning from coarse labels [10,5,11,12], e.g., C2FS [10]. Ristel et.…”

Section: Weakly-supervised Learningmentioning

confidence: 99%

Coarse-To-Fine Incremental Few-Shot Learning

Xiang¹,

Tan²,

Wan³

et al. 2021

Preprint

View full text Add to dashboard Cite

Different from fine-tuning models pre-trained on a large-scale dataset of preset classes, class-incremental learning (CIL) aims to recognize novel classes over time without forgetting pre-trained classes. However, a given model will be challenged by test images with finer-grained classes, e.g., a basenji is at most recognized as a dog. Such images form a new training set (i.e., support set) so that the incremental model is hoped to recognize a basenji (i.e., query) as a basenji next time. This paper formulates such a hybrid natural problem of coarse-to-fine fewshot (C2FS) recognition as a CIL problem named C2FSCIL, and proposes a simple, effective, and theoretically-sound strategy Knowe: to learn, normalize, and freeze a classifier's weights from fine labels, once learning an embedding space contrastively from coarse labels. Besides, as CIL aims at a stability-plasticity balance, new overall performance metrics are proposed. In that sense, on CIFAR-100, BREEDS, and tieredImageNet, Knowe outperforms all recent relevant CIL/FSCIL methods that are tailored to the new problem setting for the first time.Preprint. Under review.

show abstract

“…For instance, by showing a photograph of a stranger to a child once, he/she can rapidly identify this stranger from a pile of pictures. To acquire this ability, researchers endeavor to empower the deep models with the quick and robust learning ability in data-limited scenarios, termed few-shot learning [4,21,43,53].…”

Section: Introductionmentioning

confidence: 99%

Exploring Effective Knowledge Transfer for Few-shot Object Detection

Zhao

Liu

Wang

2022

Proceedings of the 30th ACM International Conference on Multimedia

View full text Add to dashboard Cite

Recently, few-shot object detection (FSOD) has received much attention from the community, and many methods are proposed to address this problem from a knowledge transfer perspective. Though promising results have been achieved, these methods fail to achieve shot-stable: methods that excel in low-shot regimes are likely to struggle in high-shot regimes, and vice versa. We believe this is because the primary challenge of FSOD changes when the number of shots varies. In the low-shot regime, the primary challenge is the lack of inner-class variation. In the highshot regime, as the variance approaches the real one, the main hindrance to the performance comes from misalignment between learned and true distributions. However, these two distinct issues remain unsolved in most existing FSOD methods. In this paper, we propose to overcome these challenges by exploiting rich knowledge the model has learned and effectively transferring them to the novel classes. For the low-shot regime, we propose a distribution calibration method to deal with the lack of inner-class variation problem. Meanwhile, a shift compensation method is proposed to compensate for possible distribution shift during fine-tuning. For the high-shot regime, we propose to use the knowledge learned from ImageNet as guidance for the feature learning in the finetuning stage, which will implicitly align the distributions of the novel classes. Although targeted toward different regimes, these two strategies can work together to further improve the FSOD performance. Experiments on both the VOC and COCO benchmarks show that our proposed method can significantly outperform the baseline method and produce competitive results in both low-shot settings (shot<5) and high-shot settings (shot≥5). Code is available at https://github.com/JulioZhao97/EffTrans_Fsdet.git. CCS CONCEPTS• Computing methodologies → Object detection.

show abstract

Towards Cross-Granularity Few-Shot Learning: Coarse-to-Fine Pseudo-Labeling with Visual-Semantic Meta-Embedding

Cited by 15 publications

References 53 publications

Large-scale fine-grained semantic indexing of biomedical literature based on weakly-supervised deep learning

Large-scale fine-grained semantic indexing of biomedical literature based on weakly-supervised deep learning

Coarse-To-Fine Incremental Few-Shot Learning

Exploring Effective Knowledge Transfer for Few-shot Object Detection

Contact Info

Product

Resources

About