Open-vocabulary object detection aims to train a detector capable of recognizing various novel classes. Most existing studies exploit image-level weak supervision to generate pseudo object boxes for novel class training. However, the generated ...

• Open-vocabulary object detection without using box-annotated images of novel classes.
• Better exploitation of image-level weak supervision for novel class training.
• Proposed debiased curriculum self-training for accurate pseudo-...

Source-free domain adaptation (SFDA) seeks to transfer knowledge from the source domain to the target domain by leveraging only unlabeled target data and pre-trained source model, without accessing source data. Existing approaches for SFDA ...

An impediment to training accurate classifiers in supervised learning is the scarcity of labeled data. In that respect, semi-supervised learning could help by using both labeled and unlabeled data. A specific form of semi-supervised learning is ...

• Proposes sequential retraining and pseudo-labeling in mini-batches (SRPM) for self-training (ST).
• Shows that SRPM potentially improves the performance compared to full-batch ST.
• Shows the existence of an optimal data-dependent mini-...

Deep learning (DL) has revolutionized condition monitoring (CoM) in mechanical systems by reducing manual signal processing. However, DL's industrial integration is limited due to low robustness against distribution shifts. Existing ...

Existing rumor detection models have achieved remarkable performance in fully-supervised settings. However, it is time-consuming and labor-intensive to obtain extensive labeled rumor data. To mitigate the reliance on labeled data, semi-supervised ...

• A self-training framework for semi-supervised rumor detection is proposed.
• Graph self-supervised pre-training is employed to alleviate confirmation bias.
• Self-adaptive thresholds are designed to generate reliable pseudo-labels.

Shadow detection is undergoing a rapid and remarkable development along with the wide use of deep neural networks. Benefiting from a large number of training images annotated with strong pixel-level ground-truth masks, current deep shadow ...

• A novel unsupervised deep shadow detection framework is designed.
• We design an initial pseudo label generation (IPG) module by taking advantages of the complementarity of multiple traditional unsupervised shadow detection models.
• ...

In natural language processing (NLP), zero-shot topic classification requires machines to understand the contextual meanings of texts in a downstream task without using the corresponding labeled texts for training, which is highly desirable for various ...

Medical image segmentation is fundamental in the field of medical image analysis and has wide clinical applications in disease diagnosis and surgical planning etc. Current prevalent solution is to train a deep network in a fully supervised way ...

• A novel method named PSSNet trained with multiple partially labeled datasets is proposed.
• PSSNet can simultaneously segment multiple anatomic structures and lesions.
• PSSNet achieves best on fundus image segmentation and abdominal ...

Traceability links between issues and commits record valuable information about the evolutionary history of software projects. Unfortunately, these links are often missing. While deep learning stands as the current state-of-the-art (SOTA) in ...

• Introduction of deep semi-supervised learning to traceability links recovery.
• Achieving acceptable accuracy on TLR tasks with limited labeled data.
• Empirical evaluation of labeled–unlabeled data proportion and pairing rule.

3D segmentation is a core problem in computer vision and, similarly to many other dense prediction tasks, it requires large amounts of annotated data for adequate training. However, densely labeling 3D point clouds to employ fully-supervised ...

When an object detector is deployed in a novel setting it often experiences a drop in performance. This paper studies how an embodied agent can automatically fine-tune a pre-existing object detector while exploring and acquiring images in a new ...$\mathrm{}$$\mathrm{}$$\mathrm{}$

Unsupervised video anomaly detection (UVAD) aims to detect abnormal events in videos without any annotations. It remains challenging because anomalies are rare, diverse, and usually not well-defined. Existing UVAD methods are purely data-driven ...

Camera-only Bird’s Eye View (BEV) has demonstrated great potential in environment perception in a 3D space. However, most existing studies were conducted under a supervised setup which cannot scale well while handling various new data. ...

Adversarial training is currently one of the most promising ways to achieve adversarial robustness of deep models. However, even the most sophisticated training methods is far from satisfactory, as improvement in robustness requires either ...

• An adversarial training technique using self-training is proposed.
• Consistency regularization is applied to suppress the distortion of representations in latent space.
• The proposed technique can be easily generalized to other ...

Rail surface defect detection is an essential part of railroad maintenance to prevent safety accidents. Recently, deep learning-based defect detection algorithms have shown impressive detection performance. However, the performance improvement ...

The imaging principle of seismic image is different from that of natural image. There are many problems on seismic images, such as limited resolution, complex reflection feature and strong uncertainty, which leads to significant difference in the ...

• The semi-supervised semantic segmentation method based on self-training with high-stability pseudo-labels is proposed for seismic fault identification.
• The self-training strategy Fault-Seg-ST is proposed to decouple Teacher–Student ...

This paper tackles the challenge of point-supervised temporal action detection, wherein only a single frame is annotated for each action instance in the training set. Most of the current methods, hindered by the sparse nature of annotated points, ...

Display Omitted

• An innovative point-supervised framework to capture the continuity of actions.
• A novel self-training strategy for training transformers using sparse annotations.
• A unique method for generating pseudo-labels in point-supervised ...

Traditional domain adaptive semantic segmentation methods typically assume access to source domain data during training, a paradigm known as source-access domain adaptation for semantic segmentation (SASS). To address data privacy concerns in real-...

Accurate racism classification is crucial on social media, where racist and discriminatory content can harm individuals and society. Automated racism detection requires gathering and annotating a wide range of diverse and representative data as an ...

Currently, autonomous cars are extensively studied by many institutions and companies with the aim of practical implementation. Q-learning is an algorithm of reinforcement learning, which does not require a model and can be seen as an asynchronous ...