Few-shot object segmentation with a new feature aggregation module

Liu, Kaijun; Lyu, Shujing; Shivakumara, Palaiahnakote; Lu, Yue

doi:10.1016/j.displa.2023.102459

Cited by 4 publications

(1 citation statement)

References 47 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The process of fine-tuning is as follows: (1) extraction of pertinent features from the support set and query set using the pre-trained neural network; (2) derivation of class prototypes, which are representative examples for each category in the support set; (3) computation of similarity scores between query features and prototypes; and (4) assignment of logits, indicating the likelihood of query examples belonging to each support set category. Since a small amount of data may cause instability in model training, we also explored the method of feature aggregation [ 67 , 68 , 69 ] and data augmentation [ 70 ] for multiple support sets of each category separately to create more robust category prototypes. In addition, we apply feature regularization techniques to reduce over-fitting.…”

Section: Methodsmentioning

confidence: 99%

Plant and Disease Recognition Based on PMF Pipeline Domain Adaptation Method: Using Bark Images as Meta-Dataset

Cui,

Li,

Kang

et al. 2023

Plants

View full text Add to dashboard Cite

Efficient image recognition is important in crop and forest management. However, it faces many challenges, such as the large number of plant species and diseases, the variability of plant appearance, and the scarcity of labeled data for training. To address this issue, we modified a SOTA Cross-Domain Few-shot Learning (CDFSL) method based on prototypical networks and attention mechanisms. We employed attention mechanisms to perform feature extraction and prototype generation by focusing on the most relevant parts of the images, then used prototypical networks to learn the prototype of each category and classify new instances. Finally, we demonstrated the effectiveness of the modified CDFSL method on several plant and disease recognition datasets. The results showed that the modified pipeline was able to recognize several cross-domain datasets using generic representations, and achieved up to 96.95% and 94.07% classification accuracy on datasets with the same and different domains, respectively. In addition, we visualized the experimental results, demonstrating the model’s stable transfer capability between datasets and the model’s high visual correlation with plant and disease biological characteristics. Moreover, by extending the classes of different semantics within the training dataset, our model can be generalized to other domains, which implies broad applicability.

show abstract