Graph-Collaborated Auto-Encoder Hashing for Multiview Binary Clustering

Wang, Huibing; Yao, Maojin; Jiang, Guangqi; Mi, Zetian; Fu, Xianping

doi:10.1109/tnnls.2023.3239033

Cited by 39 publications

(10 citation statements)

References 51 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Then, a novel scale-aware feature pyramid structure SA-FPN is proposed to extract rich, robust features of underwater images and improve the performance of marine object detection. Wang H et al [ 21 , 22 ] aim at minimizing the reconstruction loss between input data and binary codes for autoencoder-based hashing algorithms while ignoring the potential consistency and complementarity of multi-source data, proposes an autoencoder-based multi-view binary clustering hashing algorithm that dynamically learns an associative graph with low-rank constraints, and employs collaborative learning between the autoencoder and the associative graph to learn a unified binary code. Then, considering that most existing methods have to introduce additional clustering steps to produce the final clusters, significantly reducing the unified relationship between graph learning and clustering, a multi-view clustering based on graph collaboration is proposed.…”

Section: Related Workmentioning

confidence: 99%

LAGSwin: Local attention guided Swin-transformer for thermal infrared sports object detection

Meng,

Si,

Mao

et al. 2024

PLoS ONE

View full text Add to dashboard Cite

Compared with visible light images, thermal infrared images have poor resolution, low contrast, signal-to-noise ratio, blurred visual effects, and less information. Thermal infrared sports target detection methods relying on traditional convolutional networks capture the rich semantics in high-level features but blur the spatial details. The differences in physical information content and spatial distribution of high and low features are ignored, resulting in a mismatch between the region of interest and the target. To address these issues, we propose a local attention-guided Swin-transformer thermal infrared sports object detection method (LAGSwin) to encode sports objects’ spatial transformation and orientation information. On the one hand, Swin-transformer guided by local attention is adopted to enrich the semantic knowledge of low-level features by embedding local focus from high-level features and generating high-quality anchors while increasing the embedding of contextual information. On the other hand, an active rotation filter is employed to encode orientation information, resulting in orientation-sensitive and invariant features to reduce the inconsistency between classification and localization regression. A bidirectional criss-cross fusion strategy is adopted in the feature fusion stage to enable better interaction and embedding features of different resolutions. At last, the evaluation and verification of multiple open-source sports target datasets prove that the proposed LAGSwin detection framework has good robustness and generalization ability.

show abstract

Section: Related Workmentioning

confidence: 99%

LAGSwin: Local attention guided Swin-transformer for thermal infrared sports object detection

Meng,

Si,

Mao

et al. 2024

PLoS ONE

View full text Add to dashboard Cite

show abstract

“…Recently, attention mechanisms [14,[16][17][18][19]24] have been widely applied in the field of Re-ID to improve the performance of the model, which focuses on the information in the image that is useful for identity identification and filters out the information that is not useful for identity identification by means of an adaptive weight adjustment. AGNet [17] proposes an attention module to generate attribute masks can extract more discriminative features for category recognition.…”

Section: Attention Mechanismmentioning

confidence: 99%

“…To better distinguish vehicles with similar appearance in Fig 1, the problem of small inter-class differences in vehicle Re-ID task is solved. Many studies [14][15][16][17][18][19][20] have shown the importance of applying attention mechanisms to the vehicle Re-ID task. DGPM [14] introduced non-local attention [15] after dividing the feature map into multiple regions using rigid division to enhance attentional modeling.…”

Section: Introductionmentioning

confidence: 99%

Vehicle re-identification based on dimensional decoupling strategy and non-local relations

Li,

Pang,

Meng

2024

PLoS ONE

View full text Add to dashboard Cite

Vehicle re-identification (Re-ID) is a challenging task that aims to recognize the same vehicle across different non-overlapping cameras. Existing attention mechanism-based methods for vehicle Re-ID often suffer from significant intra-class variation and inter-class variation due to various factors such as illumination, occlusion, viewpoint, etc. In this paper, we propose a novel network architecture for vehicle Re-ID, named Dimensional Decoupling Strategy and Non-local Relationship Network (DMNR-Net), which uses three modules to extract complementary features: global feature extraction module, non-local relationship capture module(NRCM), and dimensional decoupling module (DDS). The global feature extraction module captures complete and coarse-grained features from the whole image; the NRCM module extracts saliency information from feature maps in both spatial and channel dimensions; and the DDS decouples spatial and channel features into two branches to extract fine-grained features and focus on specific subspaces. We conduct extensive experiments on two popular publicly datasets, VeRi-776 and VehicleID, to evaluate the effectiveness of our method. The experimental results show that our DMNR-Net outperforms state-of-the-art methods by a large margin on both datasets.

show abstract

“…Person search is an active research area that attempts to locate and identify query people from real-world scenes, and has gained significant attention in recent years. To date, the person's information that has been collected from multiple monitors or sensors cry out for effective search tools [1][2][3]. Compared to the person re-identification (re-id) task, this process usually contain a two-stage process, including person detection and person re-id.…”

Section: Introductionmentioning

confidence: 99%

Multi-Attention-Guided Cascading Network for End-to-End Person Search

Yang

Wang

2023

Applied Sciences

View full text Add to dashboard Cite

The key procedure is to accurately identify pedestrians in complex scenes and effectively embed features from multiple vision cues. However, it is still a limitation to coordinate two tasks in the unified framework, thus leading to high computational overhead and unsatisfactory search performance. Furthermore, most methods do not take significant clues and key features of pedestrians into consideration. To remedy these issues, we introduce a novel method named Multi-Attention-Guided Cascading Network (MGCN) in this paper. Specifically, we obtain the trusted bounding box through the detection header as the label information for post-process. Based on the end-to-end network, we demonstrate the advantages of jointly learning to construct the bounding box and attention module by maximizing the complementary information from different attention modules, which can achieve optimized person search performance. Meanwhile, by imposing an aligning module on re-id feature extracted network to locate visual clues with semantic information, which can restrain redundant background information. Extensive experimental results for the two benchmark person search datasets are provided to demonstrate that the proposed MGCN markedly outperforms the state-of-the-art baselines.

show abstract

Graph-Collaborated Auto-Encoder Hashing for Multiview Binary Clustering

Cited by 39 publications

References 51 publications

LAGSwin: Local attention guided Swin-transformer for thermal infrared sports object detection

LAGSwin: Local attention guided Swin-transformer for thermal infrared sports object detection

Vehicle re-identification based on dimensional decoupling strategy and non-local relations

Multi-Attention-Guided Cascading Network for End-to-End Person Search

Contact Info

Product

Resources

About