Group-wise Inhibition based Feature Regularization for Robust Classification

Liu, Haozhe; Wu, Haoqian; Xie, Weicheng; Liu, Feng

doi:10.1109/iccv48922.2021.00053

Cited by 12 publications

(3 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The parameters of the visual and audio encoders are randomly initialized. Besides, we perform naive K-Means algorithm [39,40] for online clustering and the cluster number #class is set to 24, while the clip length, i.e. ρ of Scene Agnostic Clip Shuffling is set to 16.…”

Section: Methodsmentioning

confidence: 99%

Scene Consistency Representation Learning for Video Scene Segmentation

Wu¹,

Chen²,

Luo³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

Section: Methodsmentioning

confidence: 99%

Scene Consistency Representation Learning for Video Scene Segmentation

Wu¹,

Chen²,

Luo³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…The recent advances in convolutional neural networks (CNNs) have yielded significant improvements to various computer vision tasks, such as image classification [14,31,39] and image generation [10,19,30]. However, recent studies [16,38] revealed that CNN based models are vulnerable to the adversarial attacks and out-of-distribution samples, which seriously limits their applications in security-critical scenarios, e.g., self-driving [40], medical imaging [33] and biometrics [28,29].…”

Section: Introductionmentioning

confidence: 99%

Robust Representation via Dynamic Feature Aggregation

Liu¹,

Ji²,

Li³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Deep convolutional neural network (CNN) based models are vulnerable to the adversarial attacks. One of the possible reasons is that the embedding space of CNN based model is sparse, resulting in a large space for the generation of adversarial samples. In this study, we propose a method, denoted as Dynamic Feature Aggregation, to compress the embedding space with a novel regularization. Particularly, the convex combination between two samples are regarded as the pivot for aggregation. In the embedding space, the selected samples are guided to be similar to the representation of the pivot. On the other side, to mitigate the trivial solution of such regularization, the last fully-connected layer of the model is replaced by an orthogonal classifier, in which the embedding codes for different classes are processed orthogonally and separately. With the regularization and orthogonal classifier, a more compact embedding space can be obtained, which accordingly improves the model robustness against adversarial attacks. An averaging accuracy of 56.91% is achieved by our method on CIFAR-10 against various attack methods, which significantly surpasses a solid baseline (Mixup) by a margin of 37.31%. More surprisingly, empirical results show that, the proposed method can also achieve the state-of-the-art performance for outof-distribution (OOD) detection, due to the learned compact feature space. An F1 score of 0.937 is achieved by the proposed method, when adopting CIFAR-10 as indistribution (ID) dataset and LSUN as OOD dataset. Code is available at https://github.com/HaozheLiu-ST/DynamicFeatureAggregation.

show abstract

“…To alleviate the workload of radiologists, an accurate automated abnormality localization system for CXRs is worthwhile to develop. With the recent advances in deep neural networks [11,6,20,7], numerous modern object detectors [8,9,10], such as FCOS [19] and Faster R-CNN [15,16], have been proposed, which can be adopted for abnormality localization. However, training these detectors often requires extensive data annotated with lesion-level bounding boxes.…”

Section: Introductionmentioning

confidence: 99%

Point Beyond Class: A Benchmark for Weakly Semi-supervised Abnormality Localization in Chest X-Rays

Liu

et al. 2022

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

Accurate abnormality localization in chest X-rays (CXR) can benefit the clinical diagnosis of various thoracic diseases. However, the lesion-level annotation can only be performed by experienced radiologists, and it is tedious and time-consuming, thus difficult to acquire. Such a situation results in a difficulty to develop a fully-supervised abnormality localization system for CXR. In this regard, we propose to train the CXR abnormality localization framework via a weakly semisupervised strategy, termed Point Beyond Class (PBC), which utilizes a small number of fully annotated CXRs with lesion-level bounding boxes and extensive weakly annotated samples by points. Such a point annotation setting can provide weakly instance-level information for abnormality localization with a marginal annotation cost. Particularly, the core idea behind our PBC is to learn a robust and accurate mapping from the point annotations to the bounding boxes against the variance of annotated points. To achieve that, a regularization term, namely multi-point consistency, is proposed, which drives the model to generate the consistent bounding box from different point annotations inside the same abnormality. Furthermore, a self-supervision, termed symmetric consistency, is also proposed to deeply exploit the useful information from the weakly annotated data for abnormality localization. Experimental results on RSNA and VinDr-CXR datasets justify the effectiveness of the proposed method. When ≤20% box-level labels are used for training, an improvement of ∼5% in mAP can be achieved by our PBC, compared to the current state-of-the-art method (i.e., Point DETR). Code is available at https://github.com/HaozheLiu-ST/Point-Beyond-Class.

show abstract

Group-wise Inhibition based Feature Regularization for Robust Classification

Cited by 12 publications

References 17 publications

Scene Consistency Representation Learning for Video Scene Segmentation

Scene Consistency Representation Learning for Video Scene Segmentation

Robust Representation via Dynamic Feature Aggregation

Point Beyond Class: A Benchmark for Weakly Semi-supervised Abnormality Localization in Chest X-Rays

Contact Info

Product

Resources

About