Abstract:Conventional detectors suffer from performance degradation when dealing with long-tailed data due to a classification bias towards the majority head categories. In this paper, we contend that the learning bias originates from two factors: 1) the unequal competition arising from the imbalanced distribution of foreground categories, and 2) the lack of sample diversity in tail categories. To tackle these issues, we introduce a unified framework called BAlanced CLassification (BACL), which enables adaptive rectification of inequalities caused by disparities in category distribution and dynamic intensification of sample diversities in a synchronized manner. Specifically, a novel foreground classification balance loss (FCBL) is developed to ameliorate the domination of head categories and shift attention to difficult-to-differentiate categories by introducing pairwise class-aware margins and auto-adjusted weight terms, respectively. This loss prevents the over-suppression of tail categories in the context of unequal competition. Moreover, we propose a dynamic feature hallucination module (FHM), which enhances the representation of tail categories in the feature space by synthesizing hallucinated samples to introduce additional data variances. In this divide-and-conquer approach, BACL sets a new state-of-the-art on the challenging LVIS benchmark with a decoupled training pipeline, surpassing vanilla Faster R-CNN with ResNet-50-FPN by 5.8% AP and 16.1% AP for overall and tail categories. Extensive experiments demonstrate that BACL consistently achieves performance improvements across various datasets with different backbones and architectures. Code and models are available at <a class="link-external link-https" href="https://github.com/Tianhao-Qi/BACL" rel="external noopener nofollow">this https URL</a>.

Feature Bias Correction: A Feature Augmentation Method for Long-tailed Recognition

Feature Re-Balancing for Long-Tailed Visual Recognition.

Contrastive Learning with Hallucinating Data for Long-Tailed Face Recognition.

Mitigating Biases in Long-Tailed Recognition Via Semantic-Guided Feature Transfer

Attentive Feature Augmentation for Long-Tailed Visual Recognition

Bt-Vmf Contrastive and Collaborative Learning for Long-Tailed Visual Recognition

Feature Cloud: Improving Deep Visual Recognition with Probabilistic Feature Augmentation

Feature-Balanced Loss for Long-Tailed Visual Recognition

Long-Tailed Time Series Classification via Feature Space Rebalancing.

Data-Centric Long-Tailed Image Recognition

Feature Space Augmentation for Long-Tailed Data

Enhanced multi-branch learning for long-tailed image recognition

A dual-branch model with inter- and intra-branch contrastive loss for long-tailed recognition

Targeted Supervised Contrastive Learning for Long-Tailed Recognition

Decoupled Contrastive Learning for Long-Tailed Recognition

Long-Tailed Object Detection for Multimodal Remote Sensing Images

FCC: Feature Clusters Compression for Long-Tailed Visual Recognition

Coarse-to-fine Knowledge Transfer Based Long-Tailed Classification Via Bilateral-Sampling Network

LCReg: Long-Tailed Image Classification with Latent Categories based Recognition

Balanced Classification: A Unified Framework for Long-Tailed Object Detection

Long-Tailed Classification Based on Dual Branch Learning Network