Equalization Loss for Long-Tailed Object Recognition

• URL: http://arxiv.org/abs/2003.05176v2
• Date: Tue, 14 Apr 2020 15:31:25 GMT
• Title: Equalization Loss for Long-Tailed Object Recognition
• Authors: Jingru Tan, Changbao Wang, Buyu Li, Quanquan Li, Wanli Ouyang, Changqing Yin, Junjie Yan
• Abstract summary: State-of-the-art object detection methods still perform poorly on large vocabulary and long-tailed datasets. We propose a simple but effective loss, named equalization loss, to tackle the problem of long-tailed rare categories. Our method achieves AP gains of 4.1% and 4.8% for the rare and common categories on the challenging LVIS benchmark.
• Score: 109.91045951333835
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Object recognition techniques using convolutional neural networks (CNN) have achieved great success. However, state-of-the-art object detection methods still perform poorly on large vocabulary and long-tailed datasets, e.g. LVIS. In this work, we analyze this problem from a novel perspective: each positive sample of one category can be seen as a negative sample for other categories, making the tail categories receive more discouraging gradients. Based on it, we propose a simple but effective loss, named equalization loss, to tackle the problem of long-tailed rare categories by simply ignoring those gradients for rare categories. The equalization loss protects the learning of rare categories from being at a disadvantage during the network parameter updating. Thus the model is capable of learning better discriminative features for objects of rare classes. Without any bells and whistles, our method achieves AP gains of 4.1% and 4.8% for the rare and common categories on the challenging LVIS benchmark, compared to the Mask R-CNN baseline. With the utilization of the effective equalization loss, we finally won the 1st place in the LVIS Challenge 2019. Code has been made available at: https: //github.com/tztztztztz/eql.detectron2

Related papers

• Escaping Saddle Points for Effective Generalization on Class-Imbalanced Data [40.64419633190104]
  We analyze the class-imbalanced learning problem by examining the loss landscape of neural networks trained with re-weighting and margin-based techniques. We find that optimization methods designed to escape from saddle points can be effectively used to improve generalization on minority classes. Using SAM results in a 6.2% increase in accuracy on the minority classes over the state-of-the-art Vector Scaling Loss, leading to an overall average increase of 4% across imbalanced datasets.
  arXiv  Detail & Related papers  (2022-12-28T14:00:44Z)
• The Equalization Losses: Gradient-Driven Training for Long-tailed Object Recognition [84.51875325962061]
  We propose a gradient-driven training mechanism to tackle the long-tail problem. We introduce a new family of gradient-driven loss functions, namely equalization losses. Our method consistently outperforms the baseline models.
  arXiv  Detail & Related papers  (2022-10-11T16:00:36Z)
• Class-Level Logit Perturbation [0.0]
  Feature perturbation and label perturbation have been proven to be useful in various deep learning approaches. New methodologies are proposed to explicitly learn to perturb logits for both single-label and multi-label classification tasks. As it only perturbs on logit, it can be used as a plug-in to fuse with any existing classification algorithms.
  arXiv  Detail & Related papers  (2022-09-13T00:49:32Z)
• Understanding Square Loss in Training Overparametrized Neural Network Classifiers [31.319145959402462]
  We contribute to the theoretical understanding of square loss in classification by systematically investigating how it performs for overparametrized neural networks. We consider two cases, according to whether classes are separable or not. In the general non-separable case, fast convergence rate is established for both misclassification rate and calibration error. The resulting margin is proven to be lower bounded away from zero, providing theoretical guarantees for robustness.
  arXiv  Detail & Related papers  (2021-12-07T12:12:30Z)
• 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)
• Forest R-CNN: Large-Vocabulary Long-Tailed Object Detection and Instance Segmentation [75.93960390191262]
  We exploit prior knowledge of the relations among object categories to cluster fine-grained classes into coarser parent classes. We propose a simple yet effective resampling method, NMS Resampling, to re-balance the data distribution. Our method, termed as Forest R-CNN, can serve as a plug-and-play module being applied to most object recognition models.
  arXiv  Detail & Related papers  (2020-08-13T03:52:37Z)
• The Devil is in Classification: A Simple Framework for Long-tail Object Detection and Instance Segmentation [93.17367076148348]
  We investigate performance drop of the state-of-the-art two-stage instance segmentation model Mask R-CNN on the recent long-tail LVIS dataset. We unveil that a major cause is the inaccurate classification of object proposals. We propose a simple calibration framework to more effectively alleviate classification head bias with a bi-level class balanced sampling approach.
  arXiv  Detail & Related papers  (2020-07-23T12:49:07Z)
• Beyond Triplet Loss: Meta Prototypical N-tuple Loss for Person Re-identification [118.72423376789062]
  We introduce a multi-class classification loss, i.e., N-tuple loss, to jointly consider multiple (N) instances for per-query optimization. With the multi-class classification incorporated, our model achieves the state-of-the-art performance on the benchmark person ReID datasets.
  arXiv  Detail & Related papers  (2020-06-08T23:34:08Z)

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.