SAM-Net: Self-Attention based Feature Matching with Spatial Transformers and Knowledge Distillation

Kelenyi, Benjamin; Domsa, Victor; Tamas, Levente

doi:10.1016/j.eswa.2023.122804

Cited by 4 publications

(1 citation statement)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The network structure is shown in Figure 5. As shown in the figure, two 1D global pooling operations are utilized to aggregate the input features along the vertical and horizontal directions, respectively, while the input feature maps come from the channel-attention-weighted feature map T se [39] and spatialattention-weighted feature map T sam [40], respectively, to form two separate directionally oriented feature maps T se, h in the H-direction and T sam, w in the W-direction sensing feature maps. These two feature maps embedded with direction-specific information are encoded as two separate attention maps, each of which captures the long-range dependencies of the input feature maps in one spatial direction, thus preserving positional information.…”

Section: Local Global Feature Coordination Enhancement Module (Lgfe)mentioning

confidence: 99%

CSINet: Channel–Spatial Fusion Networks for Asymmetric Facial Expression Recognition

Cheng,

Kong

2024

Symmetry

View full text Add to dashboard Cite

Occlusion or posture change of the face in natural scenes has typical asymmetry; however, an asymmetric face plays a key part in the lack of information available for facial expression recognition. To solve the problem of low accuracy of asymmetric facial expression recognition, this paper proposes a fusion of channel global features and a spatial local information expression recognition network called the “Channel–Spatial Integration Network” (CSINet). First, to extract the underlying detail information and deepen the network, the attention residual module with a redundant information filtering function is designed, and the backbone feature-extraction network is constituted by module stacking. Second, considering the loss of information in the local key area of face occlusion, the channel–spatial fusion structure is constructed, and the channel features and spatial features are combined to enhance the accuracy of occluded facial recognition. Finally, before the full connection layer, more local spatial information is embedded into the global channel information to capture the relationship between different channel–spatial targets, which improves the accuracy of feature expression. Experimental results on the natural scene facial expression data sets RAF-DB and FERPlus show that the recognition accuracies of the modeling approach proposed in this paper are 89.67% and 90.83%, which are 13.24% and 11.52% higher than that of the baseline network ResNet50, respectively. Compared with the latest facial expression recognition methods such as CVT, PACVT, etc., the method in this paper obtains better evaluation results of masked facial expression recognition, which provides certain theoretical and technical references for daily facial emotion analysis and human–computer interaction applications.

show abstract

Section: Local Global Feature Coordination Enhancement Module (Lgfe)mentioning

confidence: 99%

CSINet: Channel–Spatial Fusion Networks for Asymmetric Facial Expression Recognition

Cheng,

Kong

2024

Symmetry

View full text Add to dashboard Cite

show abstract

Enhancing learning on uncertain pixels in self-distillation for object segmentation

Chen,

Cao,

Zheng

et al. 2024

Complex Intell. Syst.

View full text Add to dashboard Cite

Self-distillation method guides the model learning via transferring knowledge of the model itself, which has shown the advantages in object segmentation. However, it has been proved that uncertain pixels with predicted probability close to 0.5 will restrict the model performance. The existing self-distillation methods cannot guide the model to enhance its learning ability for uncertain pixels, so the improvement is limited. To boost the student model’s learning ability for uncertain pixels, a novel self-distillation method is proposed. Firstly, the predicted probability in the current training sample and the ground truth label are fused to construct the teacher knowledge, as the current predicted information can express the performance of student models and represent the uncertainty of pixels more accurately. Secondly, a quadratic mapping function between the predicted probabilities of the teacher and student model is proposed. Theoretical analysis shows that the proposed method using the mapping function can guide the model to enhance the learning ability for uncertain pixels. Finally, the essential difference of utilizing the predicted probability of the student model in self-distillation is discussed in detail. Extensive experiments were conducted on models with convolutional neural networks and Transformer architectures as the backbone networks. The results on four public datasets demonstrate that the proposed method can effectively improve the student model performance.

show abstract

Traversability Map Estimation for Agricultural Robots

Sucală,

Tamás

2024

2024 IEEE International Conference on Automation, Quality and Testing, Robotics (AQTR)

View full text Add to dashboard Cite

SAM-Net: Self-Attention based Feature Matching with Spatial Transformers and Knowledge Distillation

Cited by 4 publications

References 33 publications

CSINet: Channel–Spatial Fusion Networks for Asymmetric Facial Expression Recognition

CSINet: Channel–Spatial Fusion Networks for Asymmetric Facial Expression Recognition

Enhancing learning on uncertain pixels in self-distillation for object segmentation

Traversability Map Estimation for Agricultural Robots

Contact Info

Product

Resources

About