Exploiting Web Images for Fine-Grained Visual Recognition via Dynamic Loss Correction and Global Sample Selection
Huafeng Liu, Haofeng Zhang, Jianfeng Lu, Zhenmin Tang · IEEE Transactions on Multimedia · 2021
To distinguish subtle differences among fine-grained categories, a large amount of well-labeled images are typically required. However, acquiring manual annotations for fine-grained categories is an extremely difficult task as it usually has a high demand for professional knowledge. To this end, directly leveraging web images for learning fine-grained models becomes a natural choice. Nevertheless, due to the existence of label noise, this learning paradigm tends to have a poor performance. In this work, we propose an end-to-end approach by combining dynamic loss correction and global sample selection to alleviate the problem of label noise. Specifically, we leverage the network to predict all samples, record the predictions of recent several epochs, and calculate the uncertainly-based dynamic loss for global sample selection. Extensive experiments on three benchmark datasets demonstrate the effectiveness of our proposed approach. The source code of our approach has been released on the website:https://github.com/NUST-Machine-Intelligence-Laboratory/dlc.