A novel FCNs‐ConvLSTM network for video salient object detection

Huang, Hai; Liu, Chang; Lan, Tian; Mu, Junsheng; Jing, Xiaojun

doi:10.1002/cta.2924

Cited by 9 publications

(4 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The difference between ConvLSTM and LSTM is that ConvLSTM takes in 3D data instead of 1D data as its input and produce 3D output vectors. ConvLSTM has been applied in change detection of hyperspectral images [24], video salient object detection [25], and Controllable Space-Time Video Super-Resolution [26]. Both LRCN and ConvLSTM are deeply explain in Section 2.5.…”

Section: Introductionmentioning

confidence: 99%

Emergency sign language recognition from variant of convolutional neural network (CNN) and long short term memory (LSTM) models

As'ari,

Sufri,

2024

Int. J. Adv. Intell. Informatics

View full text Add to dashboard Cite

Sign language is the primary communication tool used by the deaf community and people with speaking difficulties, especially during emergencies. Numerous deep learning models have been proposed to solve the sign language recognition problem. Recently. Bidirectional LSTM (BLSTM) has been proposed and used in replacement of Long Short-Term Memory (LSTM) as it may improve learning long-team dependencies as well as increase the accuracy of the model. However, there needs to be more comparison for the performance of LSTM and BLSTM in LRCN model architecture in sign language interpretation applications. Therefore, this study focused on the dense analysis of the LRCN model, including 1) training the CNN from scratch and 2) modeling with pre-trained CNN, VGG-19, and ResNet50. Other than that, the ConvLSTM model, a special variant of LSTM designed for video input, has also been modeled and compared with the LRCN in representing emergency sign language recognition. Within LRCN variants, the performance of a small CNN network was compared with pre-trained VGG-19 and ResNet50V2. A dataset of emergency Indian Sign Language with eight classes is used to train the models. The model with the best performance is the VGG-19 + LSTM model, with a testing accuracy of 96.39%. Small LRCN networks, which are 5 CNN subunits + LSTM and 4 CNN subunits + BLSTM, have 95.18% testing accuracy. This performance is on par with our best-proposed model, VGG + LSTM. By incorporating bidirectional LSTM (BLSTM) into deep learning models, the ability to understand long-term dependencies can be improved. This can enhance accuracy in reading sign language, leading to more effective communication during emergencies.

show abstract

Section: Introductionmentioning

confidence: 99%

Emergency sign language recognition from variant of convolutional neural network (CNN) and long short term memory (LSTM) models

As'ari,

Sufri,

2024

Int. J. Adv. Intell. Informatics

View full text Add to dashboard Cite

show abstract

“…Meanwhile, to improve the accuracy and robustness of power line detection, some technologies for image augmentation are also applied in enhancing the original images and reduce the disturbance of background noise as well [10]. As a matter of fact, whether the final detection performances of the above methods are useful mainly depends on the previous steps [12]. For example, the line detection and the selected edge candidates are easily interfered by the background and blurred cables of some images, which may affect the results to some extent [13].…”

Section: Introductionmentioning

confidence: 99%

An instance segmentation approach for power lines detection based on image enhancement and deep neural networks

Luo,

Yin,

et al. 2023

Second International Conference on Energy, Power, and Electrical Technology (ICEPET 2023)

View full text Add to dashboard Cite

Power lines detection (PLD) plays a key role in ensuring the convenience and stability of social urban life. Deep learning-based methods have been widely investigated for power lines detection, which can prevent the operators from damage. However, due to the influence of light, angle, position and other factors, how to realize the reliable power line detection (PLD) automatically in real time and prevent accidents is a challenging problem. In this paper, a novel framework based yolcat is proposed to enhance the images and automatically detect power lines based on deep learning. The novel framework is by design fully convolutional, and it consists of two modules: (i) an image enhancement with Gaussian Pyramid, (ii) a line segment regressor. To evaluate the proposed methods, a public dataset is utilized to test the performance of line detection with standard metric of mean average precision (mAP). Results indicate that the proposed methods show the best performance of power line detection with the image enhancement across baseline methods

show abstract

“…The video salient object detection (VSOD), also known as zero-shot video segmentation [1], [2], [3], [4], [5], [6], has received extensive research attention in recent years, whose primary objective is to segment video objects that attract the human visual attention most [7], [8], [9]. Different from the widely studied image salient object detection (ISOD) using spatial information only [10], [11], [12], the temporal information provided by the video data makes the saliency detection task more difficult [13], [14], [15], and we give an in-depth discussion regarding this issue to clearly demonstrate our motivation.…”

Section: Introductionmentioning

confidence: 99%

A Novel Long-term Iterative Mining Scheme for Video Salient Object Detection

Chen¹,

Wang²,

Fang³

et al. 2022

Preprint

View full text Add to dashboard Cite

The existing state-of-the-art (SOTA) video salient object detection (VSOD) models have widely followed short-term methodology, which dynamically determines the balance between spatial and temporal saliency fusion by solely considering the current consecutive limited frames. However, the short-term methodology has one critical limitation, which conflicts with the real mechanism of our visual system -a typical longterm methodology. As a result, failure cases keep showing up in the results of the current SOTA models, and the short-term methodology becomes the major technical bottleneck. To solve this problem, this paper proposes a novel VSOD approach, which performs VSOD in a complete long-term way. Our approach converts the sequential VSOD, a sequential task, to a data mining problem, i.e., decomposing the input video sequence to object proposals in advance and then mining salient object proposals as much as possible in an easy-to-hard way. Since all object proposals are simultaneously available, the proposed approach is a complete long-term approach, which can alleviate some difficulties rooted in conventional short-term approaches. In addition, we devised an online updating scheme that can grasp the most representative and trustworthy pattern profile of the salient objects, outputting framewise saliency maps with rich details and smoothing both spatially and temporally. The proposed approach outperforms almost all SOTA models on five widely used benchmark datasets.

show abstract

A novel FCNs‐ConvLSTM network for video salient object detection

Cited by 9 publications

References 31 publications

Emergency sign language recognition from variant of convolutional neural network (CNN) and long short term memory (LSTM) models

Emergency sign language recognition from variant of convolutional neural network (CNN) and long short term memory (LSTM) models

An instance segmentation approach for power lines detection based on image enhancement and deep neural networks

A Novel Long-term Iterative Mining Scheme for Video Salient Object Detection

Contact Info

Product

Resources

About