Leveraging Global and Background Contrast for Salient Object Detection in Hyperspectral Images

Hao, Jianhua; Li, Jianan; Pan, Chenguang; Huang, Chen; Xu, Tingfa; Cheng, Haobo

doi:10.1109/icmcce51767.2020.00474

Cited by 2 publications

(3 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…"w/o A 5 CM ". As shown in Table 6, the performance degradation of "w/o A 4 AF " validates that attention feature enhancement is beneficial for exploring the feature cues of each side-output. Besides, the necessity of cross-modality interaction can be confirmed by the comparison of "w/o A 5 CM ".…”

Section: Effectiveness Of Afe + CMImentioning

confidence: 70%

“…Some visual comparisons are shown in Figures 10 and 11, and it can be seen that after applying the cross-modality fusion module and sparse Transformer, the model can effectively compensate for the error between coarse prediction and ground truth, thus achieving more accurate results. 4 Experiments demonstrate that feature correlation scores are of critical importance for our network, which indicates that adding feature ranges and relational selection is effective for feature representation.…”

Section: Effectiveness Of Each Componentmentioning

confidence: 91%

“…Salient object detection (SOD) seeks to simulate human vision and identify the most significant objects or regions in various visual scene tasks [1]. It is one of the essential preprocessing techniques in many computer vision tasks [2], such as person re-identification [3] and image segmentation [4,5].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

SSTNet: Saliency sparse transformers network with tokenized dilation for salient object detection

Yang

Liu

Dong

et al. 2023

IET Image Processing

View full text Add to dashboard Cite

The vision Transformer structure performs better in salient object detection than the convolutional neural network (CNN)‐based approach. Vision Transformer predicts saliency by modelling long‐range dependencies from sequence to sequence with convolution‐free. It is challenging to distinguish the salient objects' location and obtain structural details for the influence of extracting irrelevant contextual information. A novel saliency sparse Transformer network is proposed to exploit sparse attention to guide saliency prediction. The convolution‐like with dilation in the token to token (T2T) module is replaced to achieve relationships in larger regions and to improve contextual information fusion. An adaptive position bias module is designed for the Vision Transformer to make position bias suitable for variable‐sized RGB images. A saliency sparse Transformer module is designed to improve the concentration of attention on the global context by selecting the Top‐k of the most relevant segments to improve the detection results further. Besides, cross‐modality to exploit the complementary RGB and depth modality fusion module (CMF) is used to take advantage of the complementary RGB image features and spatial depth information to enhance the feature fusion performance. Extensive experiments on multiple benchmark datasets demonstrate this method's effectiveness and superiority that it is suitable for saliency prediction comparable to state‐of‐the‐art RGB and RGB‐D saliency methods.

show abstract

Section: Effectiveness Of Afe + CMImentioning

confidence: 70%

Section: Effectiveness Of Each Componentmentioning

confidence: 91%

See 1 more Smart Citation

SSTNet: Saliency sparse transformers network with tokenized dilation for salient object detection

Yang

Liu

Dong

et al. 2023

IET Image Processing

View full text Add to dashboard Cite

show abstract

A Survey on Hyperspectral Image Classification and Object Detection Techniques

Bhanarkar

Amdani

2022

IJSRCSEIT

View full text Add to dashboard Cite

Machine Learning is vast field which finds its application in almost every field. The image classification is one of the important application of Supervised Machine learning algorithms. Image classification is basically concerned with identifying the objects in the images. The complexity of this task is dependent on the image features and type of images. For the research work here, the hyperspectral images are considered for deep learning based image classification. The object detection in the Hyperspectral images have applications in various areas including defense, precision agriculture, atmospheric analysis, environmental analysis, anomaly detection, fraud detection , etc. The work presented here is divided into broad survey of image classification methods using machine learning and deep learning methods. Continuing with this work, the further work presents object detection methods in ML and DL. The later work presents the deep review of the research articles over Hyperspectral image classification using Machine Learning and Deep Learning Algorithms. A lot of challenges are present to solve the object detection problems in Hyperspectral images. The later section of this work describes the object detection based on Hyperspectral images survey in detail highlighting the major developments.

show abstract

Leveraging Global and Background Contrast for Salient Object Detection in Hyperspectral Images

Cited by 2 publications

References 19 publications

SSTNet: Saliency sparse transformers network with tokenized dilation for salient object detection

SSTNet: Saliency sparse transformers network with tokenized dilation for salient object detection

A Survey on Hyperspectral Image Classification and Object Detection Techniques

Contact Info

Product

Resources

About