Camouflaged object detection aims to completely segment objects immersed in their surroundings from the background. However, existing deep learning methods often suffer from the following shortcomings: (1) They have difficulty in accurately ...

• We propose a LFNet tailored for COD tasks. It enhances COD performance by mining fine-grained features and implementing a top-down fusion approach.
• We develop a SLM to dynamically capture the structural information of targets from the ...

Traffic accident detection using video surveillance is valuable research work in intelligent transportation systems. It is useful for responding to traffic accidents promptly that can avoid traffic jam or prevent secondary accident. In traffic ...

• The CF-SOLT method includes a correlation filter to prevent vehicle ID switching.
• The traffic accident detection method is proposed by tracking occluded vehicles.
• The combination of pedestrian and vehicle behavior reduces accident ...

The recovery of 3D interacting hands meshes in the wild (ITW) is crucial for 3D full-body mesh reconstruction, especially when limited 3D annotations are available. The recent ITW interacting hands recovery method brings two hands to a shared 2D ...

• A weight-shared IoU-guided Transformer is designed to enhance the feature representation of interacting hands in the wild.
• Augmented ground truth bounding boxes are used during training to reduce the impact of detection variability, ...

Fiducial markers are a computer vision tool used for object pose estimation and detection. These markers are highly useful in fields such as industry, medicine and logistics. However, optimal lighting conditions are not always available, and ...

• A fiducial marker detection system under difficult lighting, using neural networks.
• A straightforward method to generate synthetic training data.
• A new real-world dataset featuring ArUco markers under difficult lighting conditions.

Display Omitted

The current face scanning era can quickly and conveniently attain identity authentication, but face images imply sensitive information simultaneously. Under such context, we introduce a novel cancelable face recognition methodology by using ...

• MRQSVD is introduced for pyramid representation of bimodal face images.
• Three-stream ConvNet with predefined filters is developed for features extraction.
• The proposed cancelable recognition scheme outperforms several existed ...

Despite great efforts in recent years to research robust facial landmark localization methods, occlusion remains a challenge. To tackle this challenge, we propose a model called the Landmark-in-Facial-Component Network (LFCNet). Unlike mainstream ...

Display Omitted

• We propose a new network that utilizes facial structure to address occlusion problem.
• A component localization module is designed to coarsely locate facial component.
• An offset localization module is used to draw component shapes ...

Unmanned aerial vehicle (UAV) image object detection has garnered considerable attentions in fields such as Intelligent transportation, urban management and agricultural monitoring. However, it suffers from key challenges of the deficiency in ...

• Exploiting self-attention mechanism to capture and integrate long-range dependencies.
• Using normal distribution-based prior assigner to improve accuracy of small targets.
• Using attention-guided ROI pooling to improve the quality of ...

The ability of humans to infer head poses from face shapes, and vice versa, indicates a strong correlation between them. Recent studies on face alignment used head pose information to predict facial landmarks in computer vision tasks. However, ...

Display Omitted

• We propose a novel approach that recalibrates feature maps through channel attention fusing head pose information.
• We design a dual-dimensional network with 2D and 3D kernels to address the limitations of each feature map type.
• We ...

Human-Object Interaction (HOI) detection task aims to recognize and understand interactions between humans and objects depicted in images. Unlike instance recognition tasks, which focus on isolated objects, HOI detection requires considering ...

• Our framework synergizes instance identities and visual signals, leveraging textual and visual features to identify HOIs.
• We explore the role of instance identity in HOI detection, specifying potential action categories for human-...

In recent years, Transformer-based object detection has achieved leaps and bounds in performance. Nevertheless, these methods still face some problems such as difficulty in detecting heavy occluded objects and tiny objects. Besides, the ...

• We propose an adaptive graph reasoning network for object detection.
• AGRN achieves the interaction of feature maps at various stages.
• We construct a dynamic relation graph and capture the specific relation in an image.
• AGRN ...

By defining effective supervision labels for network training, the performance of object detectors can be improved without incurring additional inference costs. Current label assignment strategies generally require two steps: first, constructing ...

• A very simple but efficient label assignment method for object detection.
• The effect of prior knowledge in label assignment is dynamically adjusted.
• A probability map is generated automatically to supervise classification.
• A ...

Multi-scale object detection is a preeminent challenge in computer vision and image processing. Several deep learning models that are designed to detect various objects miss out on the detection capabilities for small objects, reducing their ...

Display Omitted

• Improved Object Detection architecture for overhead imagery
• A novel multi-scale attention mechanism to reduce false positives
• Dilated convolutions to enhance receptive field and preserve spatial details
• Sub-pixel convolution to ...

In the field of autonomous driving, perception tasks based on Bird's-Eye-View (BEV) have attracted considerable research attention due to their numerous benefits. Despite recent advancements in performance, efficiency remains a challenge for real-...

• Efficient BEV framework for 3D object detection.
• Incorporates 2D auxiliary branch and 4D information.
• GPU memory-efficient lifting strategy.
• Prior mask evaluates feature importance at different depths.
• Tailored BEV encoder ...

The imbalanced distribution of different object categories poses a challenge for training accurate object recognition models in driving scenes. Supervised machine learning models trained on imbalanced data are biased and easily overfit the ...

• Data augmentation for driving scene object recognition using generative models.
• A systematic benchmark of generative models for lidar point clouds.
• L-GAN boosts point-based and graph-based object recognition methods effectively.

Traditional re-identification (ReID) methods heavily rely on clean and accurately annotated training data, rendering them susceptible to label noise in real-world scenarios. Although some noise-robust learning methods have been proposed and ...

• A self-consistency strategy is proposed to refine our model and avoid it overfitting to the noisy labels at the beginning of model training by mining the consistency of annotations and predictions.
• A context-consistency loss is ...

This paper presents a novel method for facial expression recognition using the proposed feature complementation and multi-scale attention model with attention fusion (FCMSA-AF). The proposed model consists of four main components: the shallow ...

• Deeper and wider model extracting diverse features at the granular level.
• Feature subsets at the left and right channels contain richer scale information.
• The correlation between two parallel paths avoids similar feature learning.

Currently, 3D object detection is a research hotspot in the field of computer vision. In this paper, we have observed that the commonly used set abstraction module retains excessive irrelevant background information during downsampling, impacting ...

• Proposes mixed sampling to enhance 3D object detection precision.
• Integrates semantic features for focused foreground point sampling.
• Introduces a module for robust feature extraction from high-quality 3D proposals.
• Achieves ...

Dense pose estimation faces hurdles due to the lack of costly precise pixel-level IUV labels. Existing methods aim to overcome it by regularizing model outputs or interpolating pseudo labels. However, conventional geometric transformations often ...

• A novel learning framework improves the model robustness with unlabeled pixels, leveraging strong and weak augmentation augmentations.
• Noise Aware Module identifies incorrectpseudo labels, adjust loss weights dynamically for stably ...

We study object detection models and observe that their respective architectures are vulnerable to image distortions such as noise, compression, blur, or snow. We propose alleviating this problem by training the models with antibodies generated ...

• Object detection models show vulnerabilities to various image distortions.
• Artificial Immune Systems (AIS)-based approach, termed AISbod, to generate “antibodies” from original training samples, referred to as antigens.
• The ...

The imbalance issues in object detection training data, such as in categories, scales, and spatial distribution, result in detection models failing to effectively fit unbalanced data. Our method aims to mitigate the performance disparity caused ...

• Developed rules to address imbalance issues in object detection training data.
• Utilized pixel-level context to selectively copy object instances.
• Leveraged global context to ensure alignment of context.