Discriminative and semantic feature selection for place recognition towards dynamic environments

Tian, Yuxin; Miao, Jinyu; Wu, Xiaofeng; Yue, Haosong; Liu, Zhong; Chen, Weihai

doi:10.1016/j.patrec.2021.11.014

“…They solved the problems when new samples appeared in the video, as the newly inputted samples should not be recognized as a disorder while the recognized behavior is within the threshold of defined normal constraints. Also, the discriminative semantic features are also extracted for fixed patterns and context-aware environment recognition by utilizing contrastive learning when the behavior analysis depends on the human interaction with the environment [104], [105].…”

Section: A Contrastive Learningmentioning

confidence: 99%

Deep Neural Networks in Video Human Action Recognition: A review

Wang¹,

Zheng²,

yang³

et al. 2023

Preprint

View full text Add to dashboard Cite

<p>Currently, spatial-temporal behavior recognition is one of the most foundational tasks of computer vision. The 2D neural networks of deep learning are built for recognizing pixel-level information such as images with RGB, RGB-D, or optical flow formats, with the current increasingly wide usage of surveillance video and more tasks related to human action recognition. There are increasing tasks requiring temporal information for frames dependency analysis. The researchers have widely studied video-based recognition rather than image-based(pixel-based) only to extract more informative elements from geometry tasks. Our current related research addresses multiple novels proposed research works and compares their advantages and disadvantages between the derived deep learning frameworks rather than machine learning frameworks. The comparison happened between existing frameworks and datasets, which are video format data only. Due to the specific properties of human actions and the increasingly wide usage of deep neural networks, we collected all research works within the last three years between 2020 to 2022. In our article, the performance of deep neural networks surpassed most of the techniques in the feature learning and extraction tasks, especially video action recognition.</p>

show abstract

“…They solved the problems when new samples appeared in the video, as the newly inputted samples should not be recognized as disorder while the recognized behavior is within the threshold of defined normal constrains. Also, the discriminative semantic features are also extracted for fixed patterns and context-aware environment recognition by utilizing contrastive learning when the behavior analysis depends on the human interaction with environment [86], [87].…”

Section: A Contrastive Learningmentioning

confidence: 99%

Deep Neural Networks in Video Human Action Recognition: A review

Wang¹,

Zheng²,

yang³

et al. 2023

Preprint

2

0

View full text Add to dashboard Cite

<p>Currently, spatial-temporal behavior recognition is one of the most foundational tasks of computer vision. The 2D neural networks of deep learning are built for recognizing pixel-level information such as images with RGB, RGB-D, or optical flow formats, with the current increasingly wide usage of surveillance video and more tasks related to human action recognition. There are increasing tasks requiring temporal information for frames dependency analysis. The researchers have widely studied video-based recognition rather than image-based(pixel-based) only to extract more informative elements from geometry tasks. Our current related research addresses multiple novels proposed research works and compares their advantages and disadvantages between the derived deep learning frameworks rather than machine learning frameworks. The comparison happened between existing frameworks and datasets, which are video format data only. Due to the specific properties of human actions and the increasingly wide usage of deep neural networks, we collected all research works within the last three years between 2020 to 2022. In our article, the performance of deep neural networks surpassed most of the techniques in the feature learning and extraction tasks, especially video action recognition.</p>

show abstract

“…However, these methods are still limited by the environment during real-time processing. A more general system is to combine laser data or image data with a neural network to recognize and label dynamic objects using semantic segmentation [ 4 , 5 ]. Compared with other target detection methods, the supervisor learning method YOLOv5 can effectively improve the efficiency and accuracy of visual identification in dynamic environments [ 6 ].…”

Section: Introductionmentioning

confidence: 99%

MoTI: A Multi-Stage Algorithm for Moving Object Identification in SLAM

Hu,

Liu,

Zhang

et al. 2023

Sensors

2

0

View full text Add to dashboard Cite

Simultaneous localization and mapping (SLAM) algorithms are widely applied in fields such as autonomous driving and target tracking. However, the effect of moving objects on localization and mapping remains a challenge in natural dynamic scenarios. To overcome this challenge, this paper proposes an algorithm for dynamic point cloud detection that fuses laser and visual identification data, the multi-stage moving object identification algorithm (MoTI). The MoTI algorithm consists of two stages: rough processing and precise processing. In the rough processing stage, a statistical method is employed to preliminarily detect dynamic points based on the range image error of the point cloud. In the precise processing stage, the radius search strategy is used to statistically test the nearest neighbor points. Next, visual identification information and point cloud registration results are fused using a method of statistics and information weighting to construct a probability model for identifying whether a point cloud cluster originates from a moving object. The algorithm is integrated into the front-end of the LOAM system, which significantly improves the localization accuracy. The MoTI algorithm is evaluated on an actual indoor dynamic environment and several KITTI datasets, and the results demonstrate its ability to accurately detect dynamic targets in the background and improve the localization accuracy of the robot.

show abstract

Discriminative and semantic feature selection for place recognition towards dynamic environments

Cited by 9 publications

References 39 publications

Deep Neural Networks in Video Human Action Recognition: A review

Deep Neural Networks in Video Human Action Recognition: A review

Deep Neural Networks in Video Human Action Recognition: A review

MoTI: A Multi-Stage Algorithm for Moving Object Identification in SLAM

Contact Info

Product

Resources

About