CQ-DINO: Mitigating Gradient Dilution via Category Queries for Vast Vocabulary Object Detection

• URL: http://arxiv.org/abs/2503.18430v2
• Date: Tue, 25 Mar 2025 07:39:46 GMT
• Title: CQ-DINO: Mitigating Gradient Dilution via Category Queries for Vast Vocabulary Object Detection
• Authors: Zhichao Sun, Huazhang Hu, Yidong Ma, Gang Liu, Nemo Chen, Xu Tang, Yao Hu, Yongchao Xu,
• Abstract summary: We propose CQ-DINO, a category query-based object detection framework.<n>CQ-DINO reformulates classification as a contrastive task between object queries and learnable category queries.<n>Experiments demonstrate that CQ-DINO achieves superior performance on the challenging V3Det benchmark.
• Score: 22.6487590600505
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: With the exponential growth of data, traditional object detection methods are increasingly struggling to handle vast vocabulary object detection tasks effectively. We analyze two key limitations of classification-based detectors: positive gradient dilution, where rare positive categories receive insufficient learning signals, and hard negative gradient dilution, where discriminative gradients are overwhelmed by numerous easy negatives. To address these challenges, we propose CQ-DINO, a category query-based object detection framework that reformulates classification as a contrastive task between object queries and learnable category queries. Our method introduces image-guided query selection, which reduces the negative space by adaptively retrieving top-K relevant categories per image via cross-attention, thereby rebalancing gradient distributions and facilitating implicit hard example mining. Furthermore, CQ-DINO flexibly integrates explicit hierarchical category relationships in structured datasets (e.g., V3Det) or learns implicit category correlations via self-attention in generic datasets (e.g., COCO). Experiments demonstrate that CQ-DINO achieves superior performance on the challenging V3Det benchmark (surpassing previous methods by 2.1% AP) while maintaining competitiveness in COCO. Our work provides a scalable solution for real-world detection systems requiring wide category coverage. The dataset and code will be publicly at https://github.com/RedAIGC/CQ-DINO.

Related papers

• GLEAN: Generalized Category Discovery with Diverse and Quality-Enhanced LLM Feedback [13.969403782532957]
  Generalized Category Discovery (GCD) aims to recognize both known and novel categories in unlabeled data.<n>Previous GCD methods face significant challenges, such as difficulty in rectifying errors for confusing instances.<n>We propose GLEAN, a unified framework for generalized category discovery.
  arXiv  Detail & Related papers  (2025-02-25T18:11:37Z)
• Hierarchical Query Classification in E-commerce Search [38.67034103433015]
  E-commerce platforms typically store and structure product information and search data in a hierarchy. Efficiently categorizing user search queries into a similar hierarchical structure is paramount in enhancing user experience on e-commerce platforms as well as news curation and academic research. The inherent complexity of hierarchical query classification is compounded by two primary challenges: (1) the pronounced class imbalance that skews towards dominant categories, and (2) the inherent brevity and ambiguity of search queries that hinder accurate classification.
  arXiv  Detail & Related papers  (2024-03-09T21:55:55Z)
• Fine-grained Recognition with Learnable Semantic Data Augmentation [68.48892326854494]
  Fine-grained image recognition is a longstanding computer vision challenge. We propose diversifying the training data at the feature-level to alleviate the discriminative region loss problem. Our method significantly improves the generalization performance on several popular classification networks.
  arXiv  Detail & Related papers  (2023-09-01T11:15:50Z)
• Dynamic Conceptional Contrastive Learning for Generalized Category Discovery [76.82327473338734]
  Generalized category discovery (GCD) aims to automatically cluster partially labeled data. Unlabeled data contain instances that are not only from known categories of the labeled data but also from novel categories. One effective way for GCD is applying self-supervised learning to learn discriminate representation for unlabeled data. We propose a Dynamic Conceptional Contrastive Learning framework, which can effectively improve clustering accuracy.
  arXiv  Detail & Related papers  (2023-03-30T14:04:39Z)
• Parametric Classification for Generalized Category Discovery: A Baseline Study [70.73212959385387]
  Generalized Category Discovery (GCD) aims to discover novel categories in unlabelled datasets using knowledge learned from labelled samples. We investigate the failure of parametric classifiers, verify the effectiveness of previous design choices when high-quality supervision is available, and identify unreliable pseudo-labels as a key problem. We propose a simple yet effective parametric classification method that benefits from entropy regularisation, achieves state-of-the-art performance on multiple GCD benchmarks and shows strong robustness to unknown class numbers.
  arXiv  Detail & Related papers  (2022-11-21T18:47:11Z)
• Fine-grained Category Discovery under Coarse-grained supervision with Hierarchical Weighted Self-contrastive Learning [37.6512548064269]
  We investigate a new practical scenario called Fine-grained Category Discovery under Coarse-grained supervision (FCDC) FCDC aims at discovering fine-grained categories with only coarse-grained labeled data, which can adapt models to categories of different granularity from known ones and reduce significant labeling cost. We propose a hierarchical weighted self-contrastive network by building a novel weighted self-contrastive module and combining it with supervised learning in a hierarchical manner.
  arXiv  Detail & Related papers  (2022-10-14T12:06:23Z)
• Classifying with Uncertain Data Envelopment Analysis [0.0]
  We propose a new classification scheme premised on the reality of imperfect data. Our model uses uncertain data envelopment analysis to define a classification's proximity to equitable efficiency. We conclude by classifying the thirty stocks in the Dow Jones Industrial average into performant tiers and by classifying prostate treatments into clinically effectual categories.
  arXiv  Detail & Related papers  (2022-09-02T13:41:19Z)
• The Overlooked Classifier in Human-Object Interaction Recognition [82.20671129356037]
  We encode the semantic correlation among classes into the classification head by initializing the weights with language embeddings of HOIs. We propose a new loss named LSE-Sign to enhance multi-label learning on a long-tailed dataset. Our simple yet effective method enables detection-free HOI classification, outperforming the state-of-the-arts that require object detection and human pose by a clear margin.
  arXiv  Detail & Related papers  (2022-03-10T23:35:00Z)
• CaT: Weakly Supervised Object Detection with Category Transfer [41.34509685442456]
  A large gap exists between fully-supervised object detection and weakly-supervised object detection. We propose a novel category transfer framework for weakly supervised object detection. Our framework achieves 63.5% mAP and 80.3% CorLoc with 5 categories overlapping between two datasets.
  arXiv  Detail & Related papers  (2021-08-17T07:59:34Z)
• Fine-Grained Visual Classification with Efficient End-to-end Localization [49.9887676289364]
  We present an efficient localization module that can be fused with a classification network in an end-to-end setup. We evaluate the new model on the three benchmark datasets CUB200-2011, Stanford Cars and FGVC-Aircraft.
  arXiv  Detail & Related papers  (2020-05-11T14:07:06Z)
• Scope Head for Accurate Localization in Object Detection [135.9979405835606]
  We propose a novel detector coined as ScopeNet, which models anchors of each location as a mutually dependent relationship. With our concise and effective design, the proposed ScopeNet achieves state-of-the-art results on COCO.
  arXiv  Detail & Related papers  (2020-05-11T04:00:09Z)

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.