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

• URL: http://arxiv.org/abs/2308.02213v1
• Date: Fri, 4 Aug 2023 09:11:07 GMT
• Title: Balanced Classification: A Unified Framework for Long-Tailed Object Detection
• Authors: Tianhao Qi, Hongtao Xie, Pandeng Li, Jiannan Ge, Yongdong Zhang
• Abstract summary: Conventional detectors suffer from performance degradation when dealing with long-tailed data due to a classification bias towards the majority head categories. We introduce a unified framework called BAlanced CLassification (BACL), which enables adaptive rectification of inequalities caused by disparities in category distribution. BACL consistently achieves performance improvements across various datasets with different backbones and architectures.
• Score: 74.94216414011326
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• 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 https://github.com/Tianhao-Qi/BACL.

Related papers

• Category-Prompt Refined Feature Learning for Long-Tailed Multi-Label Image Classification [8.139529179222844]
  Category-Prompt Refined Feature Learning (CPRFL) is a novel approach for Long-Tailed Multi-Label image Classification. CPRFL initializes category-prompts from the pretrained CLIP's embeddings and decouples category-specific visual representations. We validate the effectiveness of our method on two LTMLC benchmarks and extensive experiments demonstrate the superiority of our work over baselines.
  arXiv  Detail & Related papers  (2024-08-15T12:51:57Z)
• Dual Compensation Residual Networks for Class Imbalanced Learning [98.35401757647749]
  We propose Dual Compensation Residual Networks to better fit both tail and head classes. An important factor causing overfitting is that there is severe feature drift between training and test data on tail classes. We also propose a Residual Balanced Multi-Proxies classifier to alleviate the under-fitting issue.
  arXiv  Detail & Related papers  (2023-08-25T04:06:30Z)
• Mutual Exclusive Modulator for Long-Tailed Recognition [12.706961256329572]
  Long-tailed recognition is the task of learning high-performance classifiers given extremely imbalanced training samples between categories. We introduce a mutual exclusive modulator which can estimate the probability of an image belonging to each group. Our method achieves competitive performance compared to the state-of-the-art benchmarks.
  arXiv  Detail & Related papers  (2023-02-19T07:31:49Z)
• Semi-Supervised Semantic Segmentation via Adaptive Equalization Learning [20.66927648806676]
  We propose a novel framework for semi-supervised semantic segmentation, named adaptive equalization learning (AEL) AEL balances the training of well and badly performed categories, with a confidence bank to track category-wise performance. AEL outperforms the state-of-the-art methods by a large margin on the Cityscapes and Pascal VOC benchmarks.
  arXiv  Detail & Related papers  (2021-10-11T17:59:55Z)
• Exploring Classification Equilibrium in Long-Tailed Object Detection [29.069986049436157]
  We propose to use the mean classification score to indicate the classification accuracy for each category during training. We balance the classification via an Equilibrium Loss (EBL) and a Memory-augmented Feature Sampling (MFS) method. It improves the detection performance of tail classes by 15.6 AP, and outperforms the most recent long-tailed object detectors by more than 1 AP.
  arXiv  Detail & Related papers  (2021-08-17T08:39:04Z)
• Accuracy on the Line: On the Strong Correlation Between Out-of-Distribution and In-Distribution Generalization [89.73665256847858]
  We show that out-of-distribution performance is strongly correlated with in-distribution performance for a wide range of models and distribution shifts. Specifically, we demonstrate strong correlations between in-distribution and out-of-distribution performance on variants of CIFAR-10 & ImageNet. We also investigate cases where the correlation is weaker, for instance some synthetic distribution shifts from CIFAR-10-C and the tissue classification dataset Camelyon17-WILDS.
  arXiv  Detail & Related papers  (2021-07-09T19:48:23Z)
• No Fear of Heterogeneity: Classifier Calibration for Federated Learning with Non-IID Data [78.69828864672978]
  A central challenge in training classification models in the real-world federated system is learning with non-IID data. We propose a novel and simple algorithm called Virtual Representations (CCVR), which adjusts the classifier using virtual representations sampled from an approximated ssian mixture model. Experimental results demonstrate that CCVR state-of-the-art performance on popular federated learning benchmarks including CIFAR-10, CIFAR-100, and CINIC-10.
  arXiv  Detail & Related papers  (2021-06-09T12:02:29Z)
• Adaptive Class Suppression Loss for Long-Tail Object Detection [49.7273558444966]
  We devise a novel Adaptive Class Suppression Loss (ACSL) to improve the detection performance of tail categories. Our ACSL achieves 5.18% and 5.2% improvements with ResNet50-FPN, and sets a new state of the art.
  arXiv  Detail & Related papers  (2021-04-02T05:12:31Z)
• Seesaw Loss for Long-Tailed Instance Segmentation [131.86306953253816]
  We propose Seesaw Loss to dynamically re-balance gradients of positive and negative samples for each category. The mitigation factor reduces punishments to tail categories w.r.t. the ratio of cumulative training instances between different categories. The compensation factor increases the penalty of misclassified instances to avoid false positives of tail categories.
  arXiv  Detail & Related papers  (2020-08-23T12:44:45Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.