Lune

CVPR2023Top-tier venue

Weakly Supervised Posture Mining for Fine-Grained Classification

Zhenchao Tang, Hualin Yang, Calvin Yu-Chian Chen

Abstract

Because the subtle differences between the different sub-categories of common visual categories such as bird species, fine-grained classification has been seen as a challenging task for many years. Most previous works focus towards the features in the single discriminative region isolatedly, while neglect the connection between the different discriminative regions in the whole image. However, the relationship between different discriminative regions contains rich posture information and by adding the posture information, model can learn the behavior of the object which attribute to improve the classification performance. In this paper, we propose a novel fine-grained framework named PMRC (posture mining and reverse cross-entropy), which is able to combine with different backbones to good effect. In PMRC, we use the Deep Navigator to generate the discriminative regions from the images, and then use them to construct the graph. We aggregate the graph by message passing and get the classification results. Specifically, in order to force PMRC to learn how to mine the posture information, we design a novel training paradigm, which makes the Deep Navigator and message passing communicate and train together. In addition, we propose the reverse cross-entropy (RCE) and demomenstate that compared to the cross-entropy (CE), RCE can not only promote the accurracy of our model but also generalize to promote the accuracy of other kinds of fine-grained classification models. Experimental results on benchmark datasets confirm that PMRC can achieve state-of-the-art.

Ask Lune

Know this paper in minutes, not an afternoon.

Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.

3 top-tier papers have cited it since. Know what they found before your reviewers do.

Questions to start from

Lune can draw on

the full paper · the top-tier papers that cite it · related work

Get the answer

Free to start. No credit card required.

Lineage

  • Prompt-CAM: Making Vision Transformers Interpretable for Fine-Grained Analysis cites TransFG: A Transformer Architecture for Fine-Grained Recognition
  • Transitivity Recovering Decompositions: Interpretable and Robust Fine-Grained Relationships cites Counterfactual Attention Learning for Fine-Grained Visual Categorization and Re-identification
  • Transitivity Recovering Decompositions: Interpretable and Robust Fine-Grained Relationships cites TransFG: A Transformer Architecture for Fine-Grained Recognition
  • Counterfactual Attention Learning for Fine-Grained Visual Categorization and Re-identification cites Learning Attentive Pairwise Interaction for Fine-Grained Classification
  • TransFG: A Transformer Architecture for Fine-Grained Recognition cites Learning Attentive Pairwise Interaction for Fine-Grained Classification
Builds onCited byLeft to right by year; the more cited, the higher.

Cited by top-tier papers3

  1. Transitivity Recovering Decompositions: Interpretable and Robust Fine-Grained RelationshipsAbhra Chaudhuri, Massimiliano Mancini, Zeynep Akata, Anjan DuttaNeurIPS 2023 · 5 citations
  2. Navigating Real-World Partial Label Learning: Unveiling Fine-Grained Images with AttributesHaoran Jiang, Zhihao Sun, Yingjie TianAAAI 2024 · 2 citations
  3. Prompt-CAM: Making Vision Transformers Interpretable for Fine-Grained AnalysisArpita Chowdhury, Dipanjyoti Paul, Zheda Mai, Jianyang Gu et al.CVPR 2025

Builds on9

  1. Swin Transformer: Hierarchical Vision Transformer using Shifted WindowsZe Liu, Yutong Lin, Yue Cao, Han Hu et al.ICCV 2021 · 31,683 citations
  2. Sharpness-aware Minimization for Efficiently Improving GeneralizationPierre Foret, Ariel Kleiner, Hossein Mobahi, Behnam NeyshaburICLR 2021 · 1,861 citations
  3. TransFG: A Transformer Architecture for Fine-Grained RecognitionJu He, Jieneng Chen, Shuai Liu, Adam Kortylewski et al.AAAI 2022 · 529 citations
  4. Learning Attentive Pairwise Interaction for Fine-Grained ClassificationPeiqin Zhuang, Yali Wang, Yu QiaoAAAI 2020 · 392 citations
  5. Counterfactual Attention Learning for Fine-Grained Visual Categorization and Re-identificationYongming Rao, Guangyi Chen, Jiwen Lu, Jie ZhouICCV 2021 · 330 citations

Related papers

  1. SIM-Trans: Structure Information Modeling Transformer for Fine-grained Visual CategorizationHongbo Sun, Xiangteng He, Yuxin PengACM MM 2022 · 128 citations
  2. MAVT-FG: Multimodal Audio-Visual Transformer for Weakly-supervised Fine-Grained RecognitionXiaoyu Zhou, Xiaotong Song, Hao Wu, Jingran Zhang et al.ACM MM 2022 · 4 citations
  3. Dual Cross-Attention Learning for Fine-Grained Visual Categorization and Object Re-IdentificationHaowei Zhu, Wenjing Ke, Dong Li, Ji Liu et al.CVPR 2022 · 251 citations
  4. Multi-Part Token Transformer with Dual Contrastive Learning for Fine-grained Image ClassificationChuanming Wang, Huiyuan Fu, Huadong MaACM MM 2023 · 8 citations
  5. Revisiting Pose-Normalization for Fine-Grained Few-Shot RecognitionLuming Tang, Davis Wertheimer, Bharath HariharanCVPR 2020
Dusk over the sea between two cliffs drawn in fine vertical lines