.png){kind=icon}
Return
ProtoSimi: label correction for fine-grained visual categorization
.png){kind=icon}
OA
AI
DOI：10.1007/s10994-023-06313-0.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Deep models trained by using clean data have achieved tremendous success in fine-grained image classification. Yet, they generally suffer from significant performance degradation when encountering noisy labels. Existing approaches to handle label noise, though proved to be effective for generic object recognition, usually fail on fine-grained data. The reason is that, on fine-grained data, the category difference is subtle and the training sample size is small. Then deep models could easily overfit the noisy labels. To improve the robustness of deep models on noisy data for fine-grained visual categorization, in this paper, we propose a novel learning framework named ProtoSimi. Our method employs an adaptive label correction strategy, ensuring effective learning on limited data. Specifically, our approach considers the criteria of exploring the effectiveness of both global class-prototype and part class-prototype similarities in identifying and correcting labels of samples. We evaluate our method on three standard benchmarks of fine-grained recognition. Experimental results show that our method outperforms the existing label noisy methods by a large margin. In ablation studies, we also verify that our method is non-sensitive to hyper-parameters selection and can be integrated with other FGVC methods to increase the generalization performance.
Keywords:
Label noise
Fine-grained visual categorization
Deep learning
Journal
IF:
2.9
Papers:
2.7K
Citations:
3.4W
Organization
Cited Papers
ShareSave
ShareSaveMODELING URBAN DYNAMICS USING RANDOM FOREST: IMPLEMENTING ROC AND TOC FOR MODEL EVALUATION.png)
The International Archives of the Photogrammetry, Remote Sensing and Spatial Information SciencesOAAI
The International Archives of the Photogrammetry, Remote Sensing and Spatial Information SciencesIF0
OAAIShareSaveShareSaveShareSave
View more