Re-labeling ImageNet: from Single to Multi-Labels, from Global to Localized Labels

• URL: http://arxiv.org/abs/2101.05022v1
• Date: Wed, 13 Jan 2021 11:55:58 GMT
• Title: Re-labeling ImageNet: from Single to Multi-Labels, from Global to Localized Labels
• Authors: Sangdoo Yun, Seong Joon Oh, Byeongho Heo, Dongyoon Han, Junsuk Choe, Sanghyuk Chun
• Abstract summary: ImageNet has been arguably the most popular image classification benchmark, but it is also the one with a significant level of label noise. Recent studies have shown that many samples contain multiple classes, despite being assumed to be a single-label benchmark. We argue that the mismatch between single-label annotations and effectively multi-label images is equally, if not more, problematic in the training setup, where random crops are applied.
• Score: 34.13899937264952
• License: http://creativecommons.org/licenses/by-sa/4.0/
• Abstract: ImageNet has been arguably the most popular image classification benchmark, but it is also the one with a significant level of label noise. Recent studies have shown that many samples contain multiple classes, despite being assumed to be a single-label benchmark. They have thus proposed to turn ImageNet evaluation into a multi-label task, with exhaustive multi-label annotations per image. However, they have not fixed the training set, presumably because of a formidable annotation cost. We argue that the mismatch between single-label annotations and effectively multi-label images is equally, if not more, problematic in the training setup, where random crops are applied. With the single-label annotations, a random crop of an image may contain an entirely different object from the ground truth, introducing noisy or even incorrect supervision during training. We thus re-label the ImageNet training set with multi-labels. We address the annotation cost barrier by letting a strong image classifier, trained on an extra source of data, generate the multi-labels. We utilize the pixel-wise multi-label predictions before the final pooling layer, in order to exploit the additional location-specific supervision signals. Training on the re-labeled samples results in improved model performances across the board. ResNet-50 attains the top-1 classification accuracy of 78.9% on ImageNet with our localized multi-labels, which can be further boosted to 80.2% with the CutMix regularization. We show that the models trained with localized multi-labels also outperforms the baselines on transfer learning to object detection and instance segmentation tasks, and various robustness benchmarks. The re-labeled ImageNet training set, pre-trained weights, and the source code are available at {https://github.com/naver-ai/relabel_imagenet}.

Related papers

• Distilling Self-Supervised Vision Transformers for Weakly-Supervised Few-Shot Classification & Segmentation [58.03255076119459]
  We address the task of weakly-supervised few-shot image classification and segmentation, by leveraging a Vision Transformer (ViT) Our proposed method takes token representations from the self-supervised ViT and leverages their correlations, via self-attention, to produce classification and segmentation predictions. Experiments on Pascal-5i and COCO-20i demonstrate significant performance gains in a variety of supervision settings.
  arXiv  Detail & Related papers  (2023-07-07T06:16:43Z)
• Pseudo Labels for Single Positive Multi-Label Learning [0.0]
  Single positive multi-label (SPML) learning is a cost-effective solution, where models are trained on a single positive label per image. In this work, we propose a method to turn single positive data into fully-labeled data: Pseudo Multi-Labels.
  arXiv  Detail & Related papers  (2023-06-01T17:21:42Z)
• G2NetPL: Generic Game-Theoretic Network for Partial-Label Image Classification [14.82038002764209]
  Multi-label image classification aims to predict all possible labels in an image. Existing works on partial-label learning focus on the case where each training image is labeled with only a subset of its positive/negative labels. This paper proposes an end-to-end Generic Game-theoretic Network (G2NetPL) for partial-label learning.
  arXiv  Detail & Related papers  (2022-10-20T17:59:21Z)
• PLMCL: Partial-Label Momentum Curriculum Learning for Multi-Label Image Classification [25.451065364433028]
  Multi-label image classification aims to predict all possible labels in an image. Existing works on partial-label learning focus on the case where each training image is annotated with only a subset of its labels. This paper proposes a new partial-label setting in which only a subset of the training images are labeled, each with only one positive label, while the rest of the training images remain unlabeled.
  arXiv  Detail & Related papers  (2022-08-22T01:23:08Z)
• Semantic-Aware Representation Blending for Multi-Label Image Recognition with Partial Labels [86.17081952197788]
  We propose to blend category-specific representation across different images to transfer information of known labels to complement unknown labels. Experiments on the MS-COCO, Visual Genome, Pascal VOC 2007 datasets show that the proposed SARB framework obtains superior performance over current leading competitors.
  arXiv  Detail & Related papers  (2022-03-04T07:56:16Z)
• Multi-label Iterated Learning for Image Classification with Label Ambiguity [3.5736176624479654]
  We propose multi-label iterated learning (MILe) to incorporate the inductive biases of multi-label learning from single labels. MILe is a simple yet effective procedure that builds a multi-label description of the image by propagating binary predictions. We show that MILe is effective reducing label noise, achieving state-of-the-art performance on real-world large-scale noisy data such as WebVision.
  arXiv  Detail & Related papers  (2021-11-23T22:10:00Z)
• Mixed Supervision Learning for Whole Slide Image Classification [88.31842052998319]
  We propose a mixed supervision learning framework for super high-resolution images. During the patch training stage, this framework can make use of coarse image-level labels to refine self-supervised learning. A comprehensive strategy is proposed to suppress pixel-level false positives and false negatives.
  arXiv  Detail & Related papers  (2021-07-02T09:46:06Z)
• Multi-Label Learning from Single Positive Labels [37.17676289125165]
  Predicting all applicable labels for a given image is known as multi-label classification. We show that it is possible to approach the performance of fully labeled classifiers despite training with significantly fewer confirmed labels.
  arXiv  Detail & Related papers  (2021-06-17T17:58:04Z)
• Semantic Segmentation with Generative Models: Semi-Supervised Learning and Strong Out-of-Domain Generalization [112.68171734288237]
  We propose a novel framework for discriminative pixel-level tasks using a generative model of both images and labels. We learn a generative adversarial network that captures the joint image-label distribution and is trained efficiently using a large set of unlabeled images. We demonstrate strong in-domain performance compared to several baselines, and are the first to showcase extreme out-of-domain generalization.
  arXiv  Detail & Related papers  (2021-04-12T21:41:25Z)
• One-bit Supervision for Image Classification [121.87598671087494]
  One-bit supervision is a novel setting of learning from incomplete annotations. We propose a multi-stage training paradigm which incorporates negative label suppression into an off-the-shelf semi-supervised learning algorithm.
  arXiv  Detail & Related papers  (2020-09-14T03:06:23Z)
• Semi-supervised Learning for Few-shot Image-to-Image Translation [89.48165936436183]
  We propose a semi-supervised method for few-shot image translation, called SEMIT. Our method achieves excellent results on four different datasets using as little as 10% of the source labels.
  arXiv  Detail & Related papers  (2020-03-30T22:46:49Z)

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.