IET Computer Vision

The Coronavirus Disease 2019 (COVID‐19) epidemic has constituted a Public Health Emergency of International Concern. Chest computed tomography (CT) can help early reveal abnormalities indicative of lung disease. Thus, accurate and automatic ...

(1) The authors propose an effective classifier‐augmented generative adversarial network framework for COVID‐19 lung lesion localisation, which provides a more accurate feature map indicating the lesion regions. The proposed framework incorporating the ...

Conventional RGB‐T salient object detection treats RGB and thermal modalities equally to locate the common salient regions. However, the authors observed that the rich colour and texture information of the RGB modality makes the objects more ...

We design a mirror complementary RGB‐T SOD network (MCNet) with Transformer and CNN hybrid architecture, aiming to improve the detection performance under challenging scenes, such as low illumination and thermal crossover. The proposed model outperforms ...

Due to the huge cost of manual annotations, the labelled data may not be sufficient to train a dynamic facial expression (DFR) recogniser with good performance. To address this, the authors propose a multi‐modal pre‐training method with a pseudo‐...

To reduce the dependency on expensive manual annotations for the dynamic expression recognition (DER) task, the authors propose a multi‐modal pre‐training method with a pseudo‐label guidance mechanism to make full use of unlabelled video data for ...

Most Siamese‐based trackers use classification and regression to determine the target bounding box, which can be formulated as a linear matching process of the template and search region. However, this only takes into account the similarity of ...

We innovatively propose object tracking using semantic‐aware ensemble learning for Siamese networks. We propose for the first time a semantic tag redetection method to rescore the tracker bounding boxes and replace the inaccurate bounding boxes. image ...

Human parsing is very important in a diverse range of industrial applications. Despite the considerable progress that has been achieved, the performance of existing methods is still less than satisfactory, since these methods learn the shared ...

We propose a Region‐level Parsing Refiner (RPR) to enhance human parsing performance by the introduction of region‐level parsing learning. image image

Due to the large computational and GPUs memory cost of semantic segmentation, some works focus on designing a lite weight model to achieve a good trade‐off between computational cost and accuracy. A common method is to combined CNN and vision ...

We propose AG Self‐Attention, Enhanced Atrous Self‐Attention, and Gate Attention to fix the ignore of multi fields context and inject detailed information. image image

A technique known as domain adaptation is utilised to address classification challenges in an unlabelled target domain by leveraging labelled source domains. Previous domain adaptation approaches have predominantly focussed on global domain ...

In recent years, a considerable number of researchers have explored class‐level domain adaptation, aiming to precisely align the distribution of diverse domains. Nevertheless, existing research on class‐level alignment tends to align domain features ...

The one‐stage object detector has been widely applied in many computer vision applications due to its high detection efficiency and simple framework. However, one‐stage detectors heavily rely on Non‐maximum Suppression to remove the duplicated ...

A lightweight subnetwork, named Quality Prediction Block, is proposed by the authors to strengthen the role of localisation quality evaluation in detection confidence. It is discovered that the features of predicted bounding boxes are closely related to ...

Although social robots equipped with visual devices may leak user information, countermeasures for ensuring privacy are not readily available, making visual privacy protection problematic. In this article, a semi‐supervised learning algorithm is ...

We propose a semi‐supervised learning algorithm for visual privacy behaviour recognition based on an improved generative adversarial network for social robots (PBR‐GAN). We implement a social robot platform and architecture for visual privacy recognition ...

Although low‐rank representation (LRR)‐based subspace learning has been widely applied for feature extraction in computer vision, how to enhance the discriminability of the low‐dimensional features extracted by LRR based subspace learning ...

The manuscript constructs a robust feature extraction model named low‐rank preserving embedding regression (LRPER) for data with noises in the field of computer vision. An alternative iteration algorithm is designed to optimise the LRPER model, and its ...

The region proposal network is indispensable to two‐stage object detection methods. It generates a fixed number of proposals that are to be classified and regressed by detection heads to produce detection boxes. However, the fixed number of ...

RPN generates a fixed number of proposals images. The authors designed a module to estimate the number of objects in an image and determined the proper number of proposals according to the number, improving the performance and decreasing the FLOPS for ...

A novel video summarisation method called the Hyperbolic Graph Convolutional Network (HVSN) is proposed, which addresses the challenges of summarising edited videos and capturing the semantic consistency of video shots at different time points. ...

For user‐generated video editing, HGCN is used to extract video features and achieve good results on the dataset. Based on the attention mechanism, feature fusion technology is used to adapt to different types of videos. image image

Siamese‐based trackers have demonstrated robust performance in object tracking, while Transformers have achieved widespread success in object detection. Currently, many researchers use a hybrid structure of convolutional neural networks and ...

The authors are committed to novel approaches to solving the obscuration problem in target tracking. A full transformer tracker with state‐of‐the‐art results in multiple benchmark datasets is proposed. image image

Accurate target prediction, especially bounding box estimation, is a key problem in visual tracking. Many recently proposed trackers adopt the refinement module called IoU predictor by designing a high‐level modulation vector to achieve bounding ...

This paper proposes a novel IoU predictor (IoUNet++) for visual tracking that uses multi‐layer fused spatial template features and depthwise separable convolutional correlations to achieve more accurate bounding box estimation. The tracker improved by ...