Weakly Supervised Representation Learning with Coarse Labels

Xu, Yuanhong; Qian, Quan; Li, Hao; Jin, Rong; Hu, Juhua

doi:10.1109/iccv48922.2021.01042

Cited by 26 publications

(39 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The CBAT can filter out the unreliable target by setting a threshold. Unlike the fixed threshold in [11], [48], the CBAT can balance the diversity between different categories and adjusts adaptively by itself. Generally speaking, the fixed threshold ignores the variance in different classes.…”

Section: Class-balanced Adaptive Thresholdmentioning

confidence: 99%

Semisupervised Semantic Segmentation With Certainty-Aware Consistency Training for Remote Sensing Imagery

Guo

Wang

Xiang

et al. 2023

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

View full text Add to dashboard Cite

Semi-supervised learning is a forcible method to lessen the cost of annotation for remote sensing semantic segmentation tasks. Recent related researches indicate that consistency training is one of the most effective strategies in semisupervised learning. The core of consistency training is maintaining model outputs consistent under various perturbations. However, the current consistency training-based semi-supervised semantic segmentation frameworks lack the analysis of model uncertainty, which increases the generation of semantic ambiguity on remote sensing images. Therefore, we propose the certainty-aware consistency training (CACT) strategy to mitigate the influence of semantic ambiguity caused by model uncertainty. The certainty-aware consistency training strategy consists of two novel parts: certainty-aware consistency correction (CACC) and class-balanced adaptive threshold (CBAT). The CACC starts with generating a high-quality prediction target, then models the importance of the consistent output target and corrects the output predictions according to the certainty map, increasing the focus on reliable predictions. The CBAT uses a dynamic class-balanced adaptive threshold to filter out unreliable predictions, further reducing the impact of semantic ambiguity. Finally, considerable experimental results on the DLRSD, WHDLD, and Potsdam demonstrate that our framework has excellent performance on semi-supervised remote sensing semantic segmentation scenarios.

show abstract

Section: Class-balanced Adaptive Thresholdmentioning

confidence: 99%

Semisupervised Semantic Segmentation With Certainty-Aware Consistency Training for Remote Sensing Imagery

Guo

Wang

Xiang

et al. 2023

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

View full text Add to dashboard Cite

show abstract

“…Though not directly related to our work, we note that label granularity has been studied in many contexts other than object localization, including action recognition [46], knowledge tracing [14], animal face alignment [26], and fashion attribute recognition [23]. In the context of image classification, prior work has tackled topics like analyzing the emergence of hierarchical structure in trained classifiers [5], identifying patterns in visual concept generalization [44], and training finer-grained image classifiers using only coarse-grained labels [48,42,58,50].…”

Section: Related Workmentioning

confidence: 99%

On Label Granularity and Object Localization

Cole

Wilber

Horn

et al. 2022

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Weakly supervised object localization (WSOL) aims to learn representations that encode object location using only image-level category labels. However, many objects can be labeled at different levels of granularity. Is it an animal, a bird, or a great horned owl? Which image-level labels should we use? In this paper we study the role of label granularity in WSOL. To facilitate this investigation we introduce iNat-Loc500, a new large-scale fine-grained benchmark dataset for WSOL. Surprisingly, we find that choosing the right training label granularity provides a much larger performance boost than choosing the best WSOL algorithm. We also show that changing the label granularity can significantly improve data efficiency.

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%

“…We also use angular normalization [10] to improve their synergy. Note that m, n index samples, τ is a temperature parameter, k − m denotes the intermediate output of the m-th sample, a negative sample, in the same class with the n-th sample, a positive sample, so as to capture intra-class cues (fine cues), and reduce unnecessary noises to the subsequent fine-grained classification [11]. L Con will be small when q n is similar with k + n and different from k − m .…”

Section: Learning Embedding-weights Contrastivelymentioning

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

Weakly Supervised Representation Learning with Coarse Labels

Cited by 26 publications

References 17 publications

Semisupervised Semantic Segmentation With Certainty-Aware Consistency Training for Remote Sensing Imagery

Semisupervised Semantic Segmentation With Certainty-Aware Consistency Training for Remote Sensing Imagery

On Label Granularity and Object Localization

Coarse-To-Fine Incremental Few-Shot Learning

Contact Info

Product

Resources

About