Two-Branch Attention Learning for Fine-Grained Class Incremental Learning

Abstract

As a long-standing research area, class incremental learning (CIL) aims to effectively learn a unified classifier along with the growth of the number of classes. Due to the small inter-class variances and large intra-class variances, fine-grained visual categorization (FGVC) as a challenging visual task has not attracted enough attention in CIL. Therefore, the localization of critical regions specialized for fine-grained object recognition plays a crucial role in FGVC. Additionally, it is important to learn fine-grained features from critical regions in fine-grained CIL for the recognition of new object classes. This paper designs a network architecture named two-branch attention learning network (TBAL-Net) for fine-grained CIL. TBAL-Net can localize critical regions and learn fine-grained feature representation by a lightweight attention module. An effective training framework is proposed for fine-grained CIL by integrating TBAL-Net into an effective CIL process. This framework is tested on three popular fine-grained object datasets, including CUB-200-2011, FGVC-Aircraft, and Stanford-Car. The comparative experimental results demonstrate that the proposed framework can achieve the state-of-the-art performance on the three fine-grained object datasets.

HSCA-Net: A Hybrid Spatial-Channel Attention Network in Multi-Scale Feature Pyramid for Document Layout Analysis

“…In [31],the proposed remote sensing image defogging network consists of both encoding and decoding, and the dual self-attention module is applied to the feature enhancement of the output feature maps of the encoding stage. It improved the definition of foggy images effectively.Zhong et al [32] integrated a dual attention network composed of position attention and channel attention into the feature extraction network, which enhanced the robustness of backbone network and achieves higher accuracy in person reidentification tasks.Guo et al [33] proposed a TBAL-Net using attention mechanism to learn fine-grained feature representation, which is an effective training framework for fine-grained class incremental learning(CIL).…”

“…Considering the limitation originates from the fixed structures of CNNs, novel convolution operators were proposed to improve the learning of spatial transformations. Dilated convolutions [33] was able to aggregates contextual information from the expanded receptive field. In [35], deformable convolution was proposed to sample spatial locations with additional self-learned offsets.…”