A Multihead Continual Learning Framework for Fine-Grained Fashion Image Retrieval with Contrastive Learning and Exponential Moving Average Distillation

• URL: http://arxiv.org/abs/2603.20648v1
• Date: Sat, 21 Mar 2026 04:58:02 GMT
• Title: A Multihead Continual Learning Framework for Fine-Grained Fashion Image Retrieval with Contrastive Learning and Exponential Moving Average Distillation
• Abstract summary: Most fine-grained fashion image retrieval methods assume a static setting, requiring full retraining when new attributes appear.<n>We propose a multihead continual learning framework for fine-grained fashion image retrieval with contrastive learning and exponential moving average distillation (MCL-FIR)<n>MCL-FIR adopts a multi-head design to accommodate evolving classes across increments, reformulates triplet inputs into doublets with InfoNCE for simpler and more effective training, and employs EMA distillation for efficient knowledge transfer.
• Score: 31.097911935522674
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Most fine-grained fashion image retrieval (FIR) methods assume a static setting, requiring full retraining when new attributes appear, which is costly and impractical for dynamic scenarios. Although pretrained models support zero-shot inference, their accuracy drops without supervision, and no prior work explores class-incremental learning (CIL) for fine-grained FIR. We propose a multihead continual learning framework for fine-grained fashion image retrieval with contrastive learning and exponential moving average (EMA) distillation (MCL-FIR). MCL-FIR adopts a multi-head design to accommodate evolving classes across increments, reformulates triplet inputs into doublets with InfoNCE for simpler and more effective training, and employs EMA distillation for efficient knowledge transfer. Experiments across four datasets demonstrate that, beyond its scalability, MCL-FIR achieves a strong balance between efficiency and accuracy. It significantly outperforms CIL baselines under similar training cost, and compared with static methods, it delivers comparable performance while using only about 30% of the training cost. The source code is publicly available in https://github.com/Dr-LingXiao/MCL-FIR.

Related papers

• Pretraining Multiple Instance Learning Networks with Multi-Teacher Distillation from Pathology Slide Foundation Models [10.354992690532146]
  Multiple instance learning (MIL) has become the main paradigm for whole-slide image (WSI) analysis in computational pathology.<n>Existing MIL aggregators are still typically trained from scratch for each downstream task.<n>We propose a distillation-based pretraining framework for MIL.
  arXiv  Detail & Related papers  (2026-07-16T08:04:25Z)
• Fly-CL: A Fly-Inspired Framework for Enhancing Efficient Decorrelation and Reduced Training Time in Pre-trained Model-based Continual Representation Learning [42.58175645355976]
  Fly-CL is a bio-inspired framework compatible with a wide range of pretrained backbones.<n>We show how Fly-CL progressively resolves multicollinearity, enabling more effective similarity matching with low time complexity.
  arXiv  Detail & Related papers  (2025-10-19T15:21:50Z)
• Smooth-Distill: A Self-distillation Framework for Multitask Learning with Wearable Sensor Data [0.0]
  This paper introduces Smooth-Distill, a novel self-distillation framework designed to simultaneously perform human activity recognition (HAR) and sensor placement detection.<n>Unlike conventional distillation methods that require separate teacher and student models, the proposed framework utilizes a smoothed, historical version of the model itself as the teacher.<n> Experimental results show that Smooth-Distill consistently outperforms alternative approaches across different evaluation scenarios.
  arXiv  Detail & Related papers  (2025-06-27T06:51:51Z)
• Learning from Stochastic Teacher Representations Using Student-Guided Knowledge Distillation [64.15918654558816]
  Self-distillation (SSD) training strategy is introduced for filtering and weighting teacher representation to distill from task-relevant representations only.<n> Experimental results on real-world affective computing, wearable/biosignal datasets from the UCR Archive, the HAR dataset, and image classification datasets show that the proposed SSD method can outperform state-of-the-art methods.
  arXiv  Detail & Related papers  (2025-04-19T14:08:56Z)
• Active Data Curation Effectively Distills Large-Scale Multimodal Models [66.23057263509027]
  Knowledge distillation (KD) is the de facto standard for compressing large-scale models into smaller ones.<n>In this work we explore an alternative, yet simple approach -- active data curation as effective distillation for contrastive multimodal pretraining.<n>Our simple online batch selection method, ACID, outperforms strong KD baselines across various model-, data- and compute-configurations.
  arXiv  Detail & Related papers  (2024-11-27T18:50:15Z)
• CLIP-CID: Efficient CLIP Distillation via Cluster-Instance Discrimination [28.061239778773423]
  Contrastive Language-Image Pre-training (CLIP) has achieved excellent performance over a wide range of tasks.<n>CLIP heavily relies on a substantial corpus of pre-training data, resulting in notable consumption of computational resources.<n>We introduce CLIP-CID, a novel distillation mechanism that effectively transfers knowledge from a large vision-language foundation model to a smaller model.
  arXiv  Detail & Related papers  (2024-08-18T11:23:21Z)
• Rethinking Classifier Re-Training in Long-Tailed Recognition: A Simple Logits Retargeting Approach [102.0769560460338]
  We develop a simple logits approach (LORT) without the requirement of prior knowledge of the number of samples per class. Our method achieves state-of-the-art performance on various imbalanced datasets, including CIFAR100-LT, ImageNet-LT, and iNaturalist 2018.
  arXiv  Detail & Related papers  (2024-03-01T03:27:08Z)
• MobileCLIP: Fast Image-Text Models through Multi-Modal Reinforced Training [17.158498267947877]
  We introduce MobileCLIP, a new family of efficient image-text models optimized for runtime performance. MobileCLIP uses knowledge transfer from an image captioning model and an ensemble of strong CLIP encoders to improve the accuracy of efficient models. Our approach avoids train-time compute overhead by storing the additional knowledge in a reinforced dataset.
  arXiv  Detail & Related papers  (2023-11-28T18:55:42Z)
• Strong Baselines for Parameter Efficient Few-Shot Fine-tuning [50.83426196335385]
  Few-shot classification (FSC) entails learning novel classes given only a few examples per class after a pre-training (or meta-training) phase. Recent works have shown that simply fine-tuning a pre-trained Vision Transformer (ViT) on new test classes is a strong approach for FSC. Fine-tuning ViTs, however, is expensive in time, compute and storage. This has motivated the design of parameter efficient fine-tuning (PEFT) methods which fine-tune only a fraction of the Transformer's parameters.
  arXiv  Detail & Related papers  (2023-04-04T16:14:39Z)
• Self-Damaging Contrastive Learning [92.34124578823977]
  Unlabeled data in reality is commonly imbalanced and shows a long-tail distribution. This paper proposes a principled framework called Self-Damaging Contrastive Learning to automatically balance the representation learning without knowing the classes. Our experiments show that SDCLR significantly improves not only overall accuracies but also balancedness.
  arXiv  Detail & Related papers  (2021-06-06T00:04:49Z)
• Jigsaw Clustering for Unsupervised Visual Representation Learning [68.09280490213399]
  We propose a new jigsaw clustering pretext task in this paper. Our method makes use of information from both intra- and inter-images. It is even comparable to the contrastive learning methods when only half of training batches are used.
  arXiv  Detail & Related papers  (2021-04-01T08:09:26Z)

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.