Search

This paper addresses semi-supervised semantic segmentation by exploiting a small set of images with pixel-level annotations (strong supervisions) and a large set of images with only image-level annotations (weak supervisions). Most existing approaches aim to generate accurate pixel-level labels from weak supervisions. However, we observe that those generated labels still inevitably contain noisy labels. Motivated by this observation, we present a novel perspective and formulate this task as a problem of learning with pixel-level label noise. Existing noisy label methods, nevertheless, mainly aim at image-level tasks, which can not capture the relationship between neighboring labels in one image. Therefore, we propose a graph-based label noise detection and correction framework to deal with pixel-level noisy labels. In particular, for the generated pixel-level noisy labels from weak supervisions by Class Activation Map (CAM), we train a clean segmentation model with strong supervisions to detect the clean labels from these noisy labels according to the cross-entropy loss. Then, we adopt a superpixel-based graph to represent the relations of spatial adjacency and semantic similarity between pixels in one image. Finally we correct the noisy labels using a Graph Attention Network (GAT) supervised by detected clean labels. We comprehensively conduct experiments on PASCAL VOC 2012, PASCAL-Context, MS-COCO and Cityscapes datasets. The experimental results show that our proposed semi-supervised method achieves the state-of-the-art performances and even outperforms the fully-supervised models on PASCAL VOC 2012 and MS-COCO datasets in some cases. [ABSTRACT FROM AUTHOR]

The goal of our work is to discover dominant objects in a very general setting where only a single unlabeled image is given. This is far more challenge than typical co-localization or weakly-supervised localization tasks. To tackle this problem, we propose a simple but effective pattern mining-based method, called Object Location Mining (OLM), which exploits the advantages of data mining and feature representation of pre-trained convolutional neural networks (CNNs). Specifically, we first convert the feature maps from a pre-trained CNN model into a set of transactions, and then discovers frequent patterns from transaction database through pattern mining techniques. We observe that those discovered patterns, i.e., co-occurrence highlighted regions, typically hold appearance and spatial consistency. Motivated by this observation, we can easily discover and localize possible objects by merging relevant meaningful patterns. Extensive experiments on a variety of benchmarks demonstrate that OLM achieves competitive localization performance compared with the state-of-the-art methods. We also evaluate our approach compared with unsupervised saliency detection methods and achieves competitive results on seven benchmark datasets. Moreover, we conduct experiments on fine-grained classification to show that our proposed method can locate the entire object and parts accurately, which can benefit to improving the classification results significantly. [ABSTRACT FROM AUTHOR]

Network tomography is a promising inference technique for network topology from end-to-end measurements. In this letter, we propose a novel binary tree pruning algorithm based on t-test to infer the network topology. A binary tree topology is first inferred using the existing Agglomerative Likelihood Tree (ALT) method, and then two samples t-test is applied to prune the binary tree, thus a general tree corresponding to the real topology is obtained. A lower bound on the correctly identified probability of the proposed method is derived. Simulation results show that the pruning method based on t-test outperforms the method which prunes the binary tree using a fixed threshold. [ABSTRACT FROM PUBLISHER]

Due to severer transistor aging at nanoscale, circuit design margin becomes extremely tight for advanced technology nodes. Thus, reliability-aware circuit design is urgently needed. In this article, a new framework to perform aging-aware static timing analysis (STA) is presented for reliability analysis. The key parts of aging-aware STA flows are workload analysis and aged delay/transition calculation. For the workload analysis, a new analytical stress probability (SP) calculation model is proposed, which considers the floating effect and signal correlations. For aged delay/transition calculation, a new aging-aware model is developed, which is accessible to large industrial libraries. The results show that the proposed model achieves high accuracy in the degradation estimation and aged-path-delay calculation. Due to its high accuracy and scalability, the proposed framework is a promising solution that is compatible with commercial Electronic Design Automation (EDA) tools. [ABSTRACT FROM AUTHOR]

In this letter, the lateral trap distributions in planar and FinFET devices are experimentally studied under various bias stress conditions of hot-carrier degradation (HCD). In contrast to the traditional understanding that the generated traps are crowded near the drain region during hot-carrier stress, it is found that the peak of the trap distribution profile will gradually move closer to the source region with the increase in ${V}_{\text {ds}}$ stress. The results suggest that the electron–electron scattering and multiple vibrational excitation are important mechanisms for the lateral trap distributions during HCD stress, which is helpful for the physical understanding of HCD. [ABSTRACT FROM AUTHOR]